Spaces:
Running
Running
major update
#46
by
MINGYISU
- opened
- Scores/VLM2Vec-Qwen2-VL-V2.0-scores_report.json +1870 -0
- Scores/VLM2Vec-Qwen2-VL-V2.1-scores_report.json +1702 -0
- app.py +98 -7
- utils.py +48 -19
- utils_v2.py +149 -0
Scores/VLM2Vec-Qwen2-VL-V2.0-scores_report.json
ADDED
@@ -0,0 +1,1870 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"metadata": {
|
3 |
+
"model_name": "VLM2Vec-Qwen2-VL-V2.0",
|
4 |
+
"model_size": 2,
|
5 |
+
"embedding_dimension": null,
|
6 |
+
"max_length_tokens": null,
|
7 |
+
"model_release_date": "2025-04-01",
|
8 |
+
"data_source": "",
|
9 |
+
"url": "",
|
10 |
+
"report_generated_date": "2025-05-30T06:33:26.444844"
|
11 |
+
},
|
12 |
+
"metrics": {
|
13 |
+
"image": {
|
14 |
+
"ImageNet-1K": {
|
15 |
+
"hit@1": 0.808,
|
16 |
+
"hit@5": 0.956,
|
17 |
+
"hit@10": 0.981,
|
18 |
+
"ndcg@1": 0.808,
|
19 |
+
"ndcg@5": 0.8917355668881671,
|
20 |
+
"ndcg@10": 0.8996344961395271,
|
21 |
+
"precision@1": 0.808,
|
22 |
+
"precision@5": 0.19120000000000004,
|
23 |
+
"precision@10": 0.09810000000000002,
|
24 |
+
"recall@1": 0.808,
|
25 |
+
"recall@5": 0.956,
|
26 |
+
"recall@10": 0.981,
|
27 |
+
"f1@1": 0.808,
|
28 |
+
"f1@5": 0.3186666666666667,
|
29 |
+
"f1@10": 0.17836363636363634,
|
30 |
+
"map@1": 0.808,
|
31 |
+
"map@5": 0.8699166666666668,
|
32 |
+
"map@10": 0.8730702380952381,
|
33 |
+
"mrr@1": 0.808,
|
34 |
+
"mrr@5": 0.8699166666666668,
|
35 |
+
"mrr@10": 0.8730702380952381,
|
36 |
+
"num_pred": 1000,
|
37 |
+
"num_data": 1000
|
38 |
+
},
|
39 |
+
"N24News": {
|
40 |
+
"hit@1": 0.729,
|
41 |
+
"hit@5": 0.954,
|
42 |
+
"hit@10": 0.985,
|
43 |
+
"ndcg@1": 0.729,
|
44 |
+
"ndcg@5": 0.8542486272909989,
|
45 |
+
"ndcg@10": 0.8643817096780151,
|
46 |
+
"precision@1": 0.729,
|
47 |
+
"precision@5": 0.19080000000000003,
|
48 |
+
"precision@10": 0.09850000000000003,
|
49 |
+
"recall@1": 0.729,
|
50 |
+
"recall@5": 0.954,
|
51 |
+
"recall@10": 0.985,
|
52 |
+
"f1@1": 0.729,
|
53 |
+
"f1@5": 0.318,
|
54 |
+
"f1@10": 0.17909090909090908,
|
55 |
+
"map@1": 0.729,
|
56 |
+
"map@5": 0.82055,
|
57 |
+
"map@10": 0.8247988095238095,
|
58 |
+
"mrr@1": 0.729,
|
59 |
+
"mrr@5": 0.82055,
|
60 |
+
"mrr@10": 0.8247988095238095,
|
61 |
+
"num_pred": 1000,
|
62 |
+
"num_data": 1000
|
63 |
+
},
|
64 |
+
"HatefulMemes": {
|
65 |
+
"hit@1": 0.557,
|
66 |
+
"hit@5": 1.0,
|
67 |
+
"hit@10": 1.0,
|
68 |
+
"ndcg@1": 0.557,
|
69 |
+
"ndcg@5": 0.8365018808321558,
|
70 |
+
"ndcg@10": 0.8365018808321558,
|
71 |
+
"precision@1": 0.557,
|
72 |
+
"precision@5": 0.20000000000000004,
|
73 |
+
"precision@10": 0.10000000000000002,
|
74 |
+
"recall@1": 0.557,
|
75 |
+
"recall@5": 1.0,
|
76 |
+
"recall@10": 1.0,
|
77 |
+
"f1@1": 0.557,
|
78 |
+
"f1@5": 0.3333333333333333,
|
79 |
+
"f1@10": 0.18181818181818182,
|
80 |
+
"map@1": 0.557,
|
81 |
+
"map@5": 0.7785,
|
82 |
+
"map@10": 0.7785,
|
83 |
+
"mrr@1": 0.557,
|
84 |
+
"mrr@5": 0.7785,
|
85 |
+
"mrr@10": 0.7785,
|
86 |
+
"num_pred": 1000,
|
87 |
+
"num_data": 1000
|
88 |
+
},
|
89 |
+
"VOC2007": {
|
90 |
+
"hit@1": 0.848,
|
91 |
+
"hit@5": 0.975,
|
92 |
+
"hit@10": 0.988,
|
93 |
+
"ndcg@1": 0.848,
|
94 |
+
"ndcg@5": 0.9199061384914041,
|
95 |
+
"ndcg@10": 0.9240520704228066,
|
96 |
+
"precision@1": 0.848,
|
97 |
+
"precision@5": 0.19500000000000006,
|
98 |
+
"precision@10": 0.09880000000000001,
|
99 |
+
"recall@1": 0.848,
|
100 |
+
"recall@5": 0.975,
|
101 |
+
"recall@10": 0.988,
|
102 |
+
"f1@1": 0.848,
|
103 |
+
"f1@5": 0.325,
|
104 |
+
"f1@10": 0.17963636363636362,
|
105 |
+
"map@1": 0.848,
|
106 |
+
"map@5": 0.9011833333333334,
|
107 |
+
"map@10": 0.9028603174603175,
|
108 |
+
"mrr@1": 0.848,
|
109 |
+
"mrr@5": 0.9011833333333334,
|
110 |
+
"mrr@10": 0.9028603174603175,
|
111 |
+
"num_pred": 1000,
|
112 |
+
"num_data": 1000
|
113 |
+
},
|
114 |
+
"SUN397": {
|
115 |
+
"hit@1": 0.709,
|
116 |
+
"hit@5": 0.927,
|
117 |
+
"hit@10": 0.958,
|
118 |
+
"ndcg@1": 0.709,
|
119 |
+
"ndcg@5": 0.8312511324463461,
|
120 |
+
"ndcg@10": 0.8413545398024116,
|
121 |
+
"precision@1": 0.709,
|
122 |
+
"precision@5": 0.18540000000000004,
|
123 |
+
"precision@10": 0.09580000000000001,
|
124 |
+
"recall@1": 0.709,
|
125 |
+
"recall@5": 0.927,
|
126 |
+
"recall@10": 0.958,
|
127 |
+
"f1@1": 0.709,
|
128 |
+
"f1@5": 0.309,
|
129 |
+
"f1@10": 0.1741818181818182,
|
130 |
+
"map@1": 0.709,
|
131 |
+
"map@5": 0.7988500000000001,
|
132 |
+
"map@10": 0.8030674603174602,
|
133 |
+
"mrr@1": 0.709,
|
134 |
+
"mrr@5": 0.7988500000000001,
|
135 |
+
"mrr@10": 0.8030674603174602,
|
136 |
+
"num_pred": 1000,
|
137 |
+
"num_data": 1000
|
138 |
+
},
|
139 |
+
"Place365": {
|
140 |
+
"hit@1": 0.362,
|
141 |
+
"hit@5": 0.639,
|
142 |
+
"hit@10": 0.728,
|
143 |
+
"ndcg@1": 0.362,
|
144 |
+
"ndcg@5": 0.5106414145262992,
|
145 |
+
"ndcg@10": 0.5392146690283933,
|
146 |
+
"precision@1": 0.362,
|
147 |
+
"precision@5": 0.12779999999999997,
|
148 |
+
"precision@10": 0.07280000000000002,
|
149 |
+
"recall@1": 0.362,
|
150 |
+
"recall@5": 0.639,
|
151 |
+
"recall@10": 0.728,
|
152 |
+
"f1@1": 0.362,
|
153 |
+
"f1@5": 0.21300000000000005,
|
154 |
+
"f1@10": 0.13236363636363638,
|
155 |
+
"map@1": 0.362,
|
156 |
+
"map@5": 0.4678333333333334,
|
157 |
+
"map@10": 0.4795035714285714,
|
158 |
+
"mrr@1": 0.362,
|
159 |
+
"mrr@5": 0.4678333333333334,
|
160 |
+
"mrr@10": 0.4795035714285714,
|
161 |
+
"num_pred": 1000,
|
162 |
+
"num_data": 1000
|
163 |
+
},
|
164 |
+
"ImageNet-A": {
|
165 |
+
"hit@1": 0.473,
|
166 |
+
"hit@5": 0.705,
|
167 |
+
"hit@10": 0.789,
|
168 |
+
"ndcg@1": 0.473,
|
169 |
+
"ndcg@5": 0.5944906327072618,
|
170 |
+
"ndcg@10": 0.6220721969948751,
|
171 |
+
"precision@1": 0.473,
|
172 |
+
"precision@5": 0.141,
|
173 |
+
"precision@10": 0.07890000000000001,
|
174 |
+
"recall@1": 0.473,
|
175 |
+
"recall@5": 0.705,
|
176 |
+
"recall@10": 0.789,
|
177 |
+
"f1@1": 0.473,
|
178 |
+
"f1@5": 0.23500000000000007,
|
179 |
+
"f1@10": 0.14345454545454547,
|
180 |
+
"map@1": 0.473,
|
181 |
+
"map@5": 0.5578833333333333,
|
182 |
+
"map@10": 0.5695039682539683,
|
183 |
+
"mrr@1": 0.473,
|
184 |
+
"mrr@5": 0.5578833333333333,
|
185 |
+
"mrr@10": 0.5695039682539683,
|
186 |
+
"num_pred": 1000,
|
187 |
+
"num_data": 1000
|
188 |
+
},
|
189 |
+
"ImageNet-R": {
|
190 |
+
"hit@1": 0.893,
|
191 |
+
"hit@5": 0.968,
|
192 |
+
"hit@10": 0.98,
|
193 |
+
"ndcg@1": 0.893,
|
194 |
+
"ndcg@5": 0.9358722749979698,
|
195 |
+
"ndcg@10": 0.9397928508884185,
|
196 |
+
"precision@1": 0.893,
|
197 |
+
"precision@5": 0.19360000000000002,
|
198 |
+
"precision@10": 0.098,
|
199 |
+
"recall@1": 0.893,
|
200 |
+
"recall@5": 0.968,
|
201 |
+
"recall@10": 0.98,
|
202 |
+
"f1@1": 0.893,
|
203 |
+
"f1@5": 0.3226666666666666,
|
204 |
+
"f1@10": 0.17818181818181816,
|
205 |
+
"map@1": 0.893,
|
206 |
+
"map@5": 0.9249166666666667,
|
207 |
+
"map@10": 0.9265563492063492,
|
208 |
+
"mrr@1": 0.893,
|
209 |
+
"mrr@5": 0.9249166666666667,
|
210 |
+
"mrr@10": 0.9265563492063492,
|
211 |
+
"num_pred": 1000,
|
212 |
+
"num_data": 1000
|
213 |
+
},
|
214 |
+
"ObjectNet": {
|
215 |
+
"hit@1": 0.651,
|
216 |
+
"hit@5": 0.861,
|
217 |
+
"hit@10": 0.901,
|
218 |
+
"ndcg@1": 0.651,
|
219 |
+
"ndcg@5": 0.7677243406532726,
|
220 |
+
"ndcg@10": 0.7805651478270549,
|
221 |
+
"precision@1": 0.651,
|
222 |
+
"precision@5": 0.17220000000000002,
|
223 |
+
"precision@10": 0.09010000000000003,
|
224 |
+
"recall@1": 0.651,
|
225 |
+
"recall@5": 0.861,
|
226 |
+
"recall@10": 0.901,
|
227 |
+
"f1@1": 0.651,
|
228 |
+
"f1@5": 0.287,
|
229 |
+
"f1@10": 0.1638181818181818,
|
230 |
+
"map@1": 0.651,
|
231 |
+
"map@5": 0.7362166666666667,
|
232 |
+
"map@10": 0.7414603174603175,
|
233 |
+
"mrr@1": 0.651,
|
234 |
+
"mrr@5": 0.7362166666666667,
|
235 |
+
"mrr@10": 0.7414603174603175,
|
236 |
+
"num_pred": 1000,
|
237 |
+
"num_data": 1000
|
238 |
+
},
|
239 |
+
"Country211": {
|
240 |
+
"hit@1": 0.255,
|
241 |
+
"hit@5": 0.444,
|
242 |
+
"hit@10": 0.531,
|
243 |
+
"ndcg@1": 0.255,
|
244 |
+
"ndcg@5": 0.35293916963262145,
|
245 |
+
"ndcg@10": 0.38097997621595336,
|
246 |
+
"precision@1": 0.255,
|
247 |
+
"precision@5": 0.0888,
|
248 |
+
"precision@10": 0.053099999999999994,
|
249 |
+
"recall@1": 0.255,
|
250 |
+
"recall@5": 0.444,
|
251 |
+
"recall@10": 0.531,
|
252 |
+
"f1@1": 0.255,
|
253 |
+
"f1@5": 0.14800000000000002,
|
254 |
+
"f1@10": 0.09654545454545456,
|
255 |
+
"map@1": 0.255,
|
256 |
+
"map@5": 0.3228666666666667,
|
257 |
+
"map@10": 0.33438412698412695,
|
258 |
+
"mrr@1": 0.255,
|
259 |
+
"mrr@5": 0.3228666666666667,
|
260 |
+
"mrr@10": 0.33438412698412695,
|
261 |
+
"num_pred": 1000,
|
262 |
+
"num_data": 1000
|
263 |
+
},
|
264 |
+
"OK-VQA": {
|
265 |
+
"hit@1": 0.517,
|
266 |
+
"hit@5": 0.779,
|
267 |
+
"hit@10": 0.841,
|
268 |
+
"ndcg@1": 0.517,
|
269 |
+
"ndcg@5": 0.6577545355945692,
|
270 |
+
"ndcg@10": 0.6778864681057628,
|
271 |
+
"precision@1": 0.517,
|
272 |
+
"precision@5": 0.15580000000000002,
|
273 |
+
"precision@10": 0.08410000000000001,
|
274 |
+
"recall@1": 0.517,
|
275 |
+
"recall@5": 0.779,
|
276 |
+
"recall@10": 0.841,
|
277 |
+
"f1@1": 0.517,
|
278 |
+
"f1@5": 0.25966666666666677,
|
279 |
+
"f1@10": 0.15290909090909094,
|
280 |
+
"map@1": 0.517,
|
281 |
+
"map@5": 0.6172833333333333,
|
282 |
+
"map@10": 0.6256357142857144,
|
283 |
+
"mrr@1": 0.517,
|
284 |
+
"mrr@5": 0.6172833333333333,
|
285 |
+
"mrr@10": 0.6256357142857144,
|
286 |
+
"num_pred": 1000,
|
287 |
+
"num_data": 1000
|
288 |
+
},
|
289 |
+
"A-OKVQA": {
|
290 |
+
"hit@1": 0.441,
|
291 |
+
"hit@5": 0.728,
|
292 |
+
"hit@10": 0.803,
|
293 |
+
"ndcg@1": 0.441,
|
294 |
+
"ndcg@5": 0.5967422009124269,
|
295 |
+
"ndcg@10": 0.6212347128491258,
|
296 |
+
"precision@1": 0.441,
|
297 |
+
"precision@5": 0.1456,
|
298 |
+
"precision@10": 0.0803,
|
299 |
+
"recall@1": 0.441,
|
300 |
+
"recall@5": 0.728,
|
301 |
+
"recall@10": 0.803,
|
302 |
+
"f1@1": 0.441,
|
303 |
+
"f1@5": 0.24266666666666672,
|
304 |
+
"f1@10": 0.146,
|
305 |
+
"map@1": 0.441,
|
306 |
+
"map@5": 0.5527833333333333,
|
307 |
+
"map@10": 0.5630289682539683,
|
308 |
+
"mrr@1": 0.441,
|
309 |
+
"mrr@5": 0.5527833333333333,
|
310 |
+
"mrr@10": 0.5630289682539683,
|
311 |
+
"num_pred": 1000,
|
312 |
+
"num_data": 1000
|
313 |
+
},
|
314 |
+
"DocVQA": {
|
315 |
+
"hit@1": 0.903,
|
316 |
+
"hit@5": 0.972,
|
317 |
+
"hit@10": 0.98,
|
318 |
+
"ndcg@1": 0.903,
|
319 |
+
"ndcg@5": 0.9423336632220655,
|
320 |
+
"ndcg@10": 0.9449923796053714,
|
321 |
+
"precision@1": 0.903,
|
322 |
+
"precision@5": 0.19440000000000004,
|
323 |
+
"precision@10": 0.09800000000000002,
|
324 |
+
"recall@1": 0.903,
|
325 |
+
"recall@5": 0.972,
|
326 |
+
"recall@10": 0.98,
|
327 |
+
"f1@1": 0.903,
|
328 |
+
"f1@5": 0.324,
|
329 |
+
"f1@10": 0.17818181818181816,
|
330 |
+
"map@1": 0.903,
|
331 |
+
"map@5": 0.9322333333333334,
|
332 |
+
"map@10": 0.9333730158730159,
|
333 |
+
"mrr@1": 0.903,
|
334 |
+
"mrr@5": 0.9322333333333334,
|
335 |
+
"mrr@10": 0.9333730158730159,
|
336 |
+
"num_pred": 1000,
|
337 |
+
"num_data": 1000
|
338 |
+
},
|
339 |
+
"InfographicsVQA": {
|
340 |
+
"hit@1": 0.591,
|
341 |
+
"hit@5": 0.809,
|
342 |
+
"hit@10": 0.867,
|
343 |
+
"ndcg@1": 0.591,
|
344 |
+
"ndcg@5": 0.7077964996199817,
|
345 |
+
"ndcg@10": 0.7268065366746665,
|
346 |
+
"precision@1": 0.591,
|
347 |
+
"precision@5": 0.16179999999999997,
|
348 |
+
"precision@10": 0.08669999999999999,
|
349 |
+
"recall@1": 0.591,
|
350 |
+
"recall@5": 0.809,
|
351 |
+
"recall@10": 0.867,
|
352 |
+
"f1@1": 0.591,
|
353 |
+
"f1@5": 0.26966666666666667,
|
354 |
+
"f1@10": 0.15763636363636363,
|
355 |
+
"map@1": 0.591,
|
356 |
+
"map@5": 0.6740666666666666,
|
357 |
+
"map@10": 0.6820595238095238,
|
358 |
+
"mrr@1": 0.591,
|
359 |
+
"mrr@5": 0.6740666666666666,
|
360 |
+
"mrr@10": 0.6820595238095238,
|
361 |
+
"num_pred": 1000,
|
362 |
+
"num_data": 1000
|
363 |
+
},
|
364 |
+
"ChartQA": {
|
365 |
+
"hit@1": 0.48,
|
366 |
+
"hit@5": 0.676,
|
367 |
+
"hit@10": 0.751,
|
368 |
+
"ndcg@1": 0.48,
|
369 |
+
"ndcg@5": 0.5851856746029632,
|
370 |
+
"ndcg@10": 0.6094499633387991,
|
371 |
+
"precision@1": 0.48,
|
372 |
+
"precision@5": 0.13520000000000001,
|
373 |
+
"precision@10": 0.07510000000000001,
|
374 |
+
"recall@1": 0.48,
|
375 |
+
"recall@5": 0.676,
|
376 |
+
"recall@10": 0.751,
|
377 |
+
"f1@1": 0.48,
|
378 |
+
"f1@5": 0.22533333333333336,
|
379 |
+
"f1@10": 0.13654545454545453,
|
380 |
+
"map@1": 0.48,
|
381 |
+
"map@5": 0.55495,
|
382 |
+
"map@10": 0.5649611111111111,
|
383 |
+
"mrr@1": 0.48,
|
384 |
+
"mrr@5": 0.55495,
|
385 |
+
"mrr@10": 0.5649611111111111,
|
386 |
+
"num_pred": 1000,
|
387 |
+
"num_data": 1000
|
388 |
+
},
|
389 |
+
"Visual7W": {
|
390 |
+
"hit@1": 0.526,
|
391 |
+
"hit@5": 0.877,
|
392 |
+
"hit@10": 0.953,
|
393 |
+
"ndcg@1": 0.526,
|
394 |
+
"ndcg@5": 0.7196026298030704,
|
395 |
+
"ndcg@10": 0.7446210533716667,
|
396 |
+
"precision@1": 0.526,
|
397 |
+
"precision@5": 0.1754,
|
398 |
+
"precision@10": 0.09530000000000001,
|
399 |
+
"recall@1": 0.526,
|
400 |
+
"recall@5": 0.877,
|
401 |
+
"recall@10": 0.953,
|
402 |
+
"f1@1": 0.526,
|
403 |
+
"f1@5": 0.2923333333333334,
|
404 |
+
"f1@10": 0.17327272727272724,
|
405 |
+
"map@1": 0.526,
|
406 |
+
"map@5": 0.6666333333333334,
|
407 |
+
"map@10": 0.6772194444444445,
|
408 |
+
"mrr@1": 0.526,
|
409 |
+
"mrr@5": 0.6666333333333334,
|
410 |
+
"mrr@10": 0.6772194444444445,
|
411 |
+
"num_pred": 1000,
|
412 |
+
"num_data": 1000
|
413 |
+
},
|
414 |
+
"ScienceQA": {
|
415 |
+
"hit@1": 0.382,
|
416 |
+
"hit@5": 0.656,
|
417 |
+
"hit@10": 0.774,
|
418 |
+
"ndcg@1": 0.382,
|
419 |
+
"ndcg@5": 0.5252539659457145,
|
420 |
+
"ndcg@10": 0.563294943869286,
|
421 |
+
"precision@1": 0.382,
|
422 |
+
"precision@5": 0.13119999999999998,
|
423 |
+
"precision@10": 0.07740000000000001,
|
424 |
+
"recall@1": 0.382,
|
425 |
+
"recall@5": 0.656,
|
426 |
+
"recall@10": 0.774,
|
427 |
+
"f1@1": 0.382,
|
428 |
+
"f1@5": 0.21866666666666673,
|
429 |
+
"f1@10": 0.14072727272727276,
|
430 |
+
"map@1": 0.382,
|
431 |
+
"map@5": 0.48206666666666664,
|
432 |
+
"map@10": 0.4976932539682539,
|
433 |
+
"mrr@1": 0.382,
|
434 |
+
"mrr@5": 0.48206666666666664,
|
435 |
+
"mrr@10": 0.4976932539682539,
|
436 |
+
"num_pred": 1000,
|
437 |
+
"num_data": 1000
|
438 |
+
},
|
439 |
+
"VizWiz": {
|
440 |
+
"hit@1": 0.434,
|
441 |
+
"hit@5": 0.613,
|
442 |
+
"hit@10": 0.672,
|
443 |
+
"ndcg@1": 0.434,
|
444 |
+
"ndcg@5": 0.5310329640421506,
|
445 |
+
"ndcg@10": 0.5504145410550396,
|
446 |
+
"precision@1": 0.434,
|
447 |
+
"precision@5": 0.12259999999999999,
|
448 |
+
"precision@10": 0.06720000000000001,
|
449 |
+
"recall@1": 0.434,
|
450 |
+
"recall@5": 0.613,
|
451 |
+
"recall@10": 0.672,
|
452 |
+
"f1@1": 0.434,
|
453 |
+
"f1@5": 0.2043333333333334,
|
454 |
+
"f1@10": 0.12218181818181818,
|
455 |
+
"map@1": 0.434,
|
456 |
+
"map@5": 0.5036,
|
457 |
+
"map@10": 0.5117742063492063,
|
458 |
+
"mrr@1": 0.434,
|
459 |
+
"mrr@5": 0.5036,
|
460 |
+
"mrr@10": 0.5117742063492063,
|
461 |
+
"num_pred": 1000,
|
462 |
+
"num_data": 1000
|
463 |
+
},
|
464 |
+
"GQA": {
|
465 |
+
"hit@1": 0.652,
|
466 |
+
"hit@5": 0.885,
|
467 |
+
"hit@10": 0.93,
|
468 |
+
"ndcg@1": 0.652,
|
469 |
+
"ndcg@5": 0.7806290021458085,
|
470 |
+
"ndcg@10": 0.7952855256833086,
|
471 |
+
"precision@1": 0.652,
|
472 |
+
"precision@5": 0.177,
|
473 |
+
"precision@10": 0.09300000000000001,
|
474 |
+
"recall@1": 0.652,
|
475 |
+
"recall@5": 0.885,
|
476 |
+
"recall@10": 0.93,
|
477 |
+
"f1@1": 0.652,
|
478 |
+
"f1@5": 0.295,
|
479 |
+
"f1@10": 0.1690909090909091,
|
480 |
+
"map@1": 0.652,
|
481 |
+
"map@5": 0.7454666666666667,
|
482 |
+
"map@10": 0.7515809523809525,
|
483 |
+
"mrr@1": 0.652,
|
484 |
+
"mrr@5": 0.7454666666666667,
|
485 |
+
"mrr@10": 0.7515809523809525,
|
486 |
+
"num_pred": 1000,
|
487 |
+
"num_data": 1000
|
488 |
+
},
|
489 |
+
"TextVQA": {
|
490 |
+
"hit@1": 0.717,
|
491 |
+
"hit@5": 0.832,
|
492 |
+
"hit@10": 0.863,
|
493 |
+
"ndcg@1": 0.717,
|
494 |
+
"ndcg@5": 0.7798532787086642,
|
495 |
+
"ndcg@10": 0.7899769316117858,
|
496 |
+
"precision@1": 0.717,
|
497 |
+
"precision@5": 0.1664,
|
498 |
+
"precision@10": 0.08630000000000002,
|
499 |
+
"recall@1": 0.717,
|
500 |
+
"recall@5": 0.832,
|
501 |
+
"recall@10": 0.863,
|
502 |
+
"f1@1": 0.717,
|
503 |
+
"f1@5": 0.2773333333333334,
|
504 |
+
"f1@10": 0.15690909090909094,
|
505 |
+
"map@1": 0.717,
|
506 |
+
"map@5": 0.7623333333333334,
|
507 |
+
"map@10": 0.7665742063492064,
|
508 |
+
"mrr@1": 0.717,
|
509 |
+
"mrr@5": 0.7623333333333334,
|
510 |
+
"mrr@10": 0.7665742063492064,
|
511 |
+
"num_pred": 1000,
|
512 |
+
"num_data": 1000
|
513 |
+
},
|
514 |
+
"VisDial": {
|
515 |
+
"hit@1": 0.827,
|
516 |
+
"hit@5": 0.968,
|
517 |
+
"hit@10": 0.986,
|
518 |
+
"ndcg@1": 0.827,
|
519 |
+
"ndcg@5": 0.9045672377369148,
|
520 |
+
"ndcg@10": 0.9104628927774298,
|
521 |
+
"precision@1": 0.827,
|
522 |
+
"precision@5": 0.19360000000000002,
|
523 |
+
"precision@10": 0.09860000000000002,
|
524 |
+
"recall@1": 0.827,
|
525 |
+
"recall@5": 0.968,
|
526 |
+
"recall@10": 0.986,
|
527 |
+
"f1@1": 0.827,
|
528 |
+
"f1@5": 0.3226666666666667,
|
529 |
+
"f1@10": 0.17927272727272725,
|
530 |
+
"map@1": 0.827,
|
531 |
+
"map@5": 0.88325,
|
532 |
+
"map@10": 0.8857265873015872,
|
533 |
+
"mrr@1": 0.827,
|
534 |
+
"mrr@5": 0.88325,
|
535 |
+
"mrr@10": 0.8857265873015872,
|
536 |
+
"num_pred": 1000,
|
537 |
+
"num_data": 1000
|
538 |
+
},
|
539 |
+
"CIRR": {
|
540 |
+
"hit@1": 0.577,
|
541 |
+
"hit@5": 0.898,
|
542 |
+
"hit@10": 0.952,
|
543 |
+
"ndcg@1": 0.577,
|
544 |
+
"ndcg@5": 0.7558069865262556,
|
545 |
+
"ndcg@10": 0.7735209989825035,
|
546 |
+
"precision@1": 0.577,
|
547 |
+
"precision@5": 0.1796,
|
548 |
+
"precision@10": 0.09520000000000002,
|
549 |
+
"recall@1": 0.577,
|
550 |
+
"recall@5": 0.898,
|
551 |
+
"recall@10": 0.952,
|
552 |
+
"f1@1": 0.577,
|
553 |
+
"f1@5": 0.2993333333333334,
|
554 |
+
"f1@10": 0.1730909090909091,
|
555 |
+
"map@1": 0.577,
|
556 |
+
"map@5": 0.7077833333333333,
|
557 |
+
"map@10": 0.7152424603174603,
|
558 |
+
"mrr@1": 0.577,
|
559 |
+
"mrr@5": 0.7077833333333333,
|
560 |
+
"mrr@10": 0.7152424603174603,
|
561 |
+
"num_pred": 1000,
|
562 |
+
"num_data": 1000
|
563 |
+
},
|
564 |
+
"VisualNews_t2i": {
|
565 |
+
"hit@1": 0.744,
|
566 |
+
"hit@5": 0.894,
|
567 |
+
"hit@10": 0.927,
|
568 |
+
"ndcg@1": 0.744,
|
569 |
+
"ndcg@5": 0.827463641275362,
|
570 |
+
"ndcg@10": 0.8383042912343103,
|
571 |
+
"precision@1": 0.744,
|
572 |
+
"precision@5": 0.17880000000000001,
|
573 |
+
"precision@10": 0.0927,
|
574 |
+
"recall@1": 0.744,
|
575 |
+
"recall@5": 0.894,
|
576 |
+
"recall@10": 0.927,
|
577 |
+
"f1@1": 0.744,
|
578 |
+
"f1@5": 0.29800000000000004,
|
579 |
+
"f1@10": 0.16854545454545453,
|
580 |
+
"map@1": 0.744,
|
581 |
+
"map@5": 0.8050333333333333,
|
582 |
+
"map@10": 0.8096071428571429,
|
583 |
+
"mrr@1": 0.744,
|
584 |
+
"mrr@5": 0.8050333333333333,
|
585 |
+
"mrr@10": 0.8096071428571429,
|
586 |
+
"num_pred": 1000,
|
587 |
+
"num_data": 1000
|
588 |
+
},
|
589 |
+
"VisualNews_i2t": {
|
590 |
+
"hit@1": 0.783,
|
591 |
+
"hit@5": 0.904,
|
592 |
+
"hit@10": 0.935,
|
593 |
+
"ndcg@1": 0.783,
|
594 |
+
"ndcg@5": 0.8479824009858002,
|
595 |
+
"ndcg@10": 0.8581904833957,
|
596 |
+
"precision@1": 0.783,
|
597 |
+
"precision@5": 0.18080000000000002,
|
598 |
+
"precision@10": 0.09350000000000001,
|
599 |
+
"recall@1": 0.783,
|
600 |
+
"recall@5": 0.904,
|
601 |
+
"recall@10": 0.935,
|
602 |
+
"f1@1": 0.783,
|
603 |
+
"f1@5": 0.3013333333333334,
|
604 |
+
"f1@10": 0.17,
|
605 |
+
"map@1": 0.783,
|
606 |
+
"map@5": 0.8292666666666666,
|
607 |
+
"map@10": 0.8335880952380953,
|
608 |
+
"mrr@1": 0.783,
|
609 |
+
"mrr@5": 0.8292666666666666,
|
610 |
+
"mrr@10": 0.8335880952380953,
|
611 |
+
"num_pred": 1000,
|
612 |
+
"num_data": 1000
|
613 |
+
},
|
614 |
+
"MSCOCO_t2i": {
|
615 |
+
"hit@1": 0.753,
|
616 |
+
"hit@5": 0.949,
|
617 |
+
"hit@10": 0.981,
|
618 |
+
"ndcg@1": 0.753,
|
619 |
+
"ndcg@5": 0.8611598177428067,
|
620 |
+
"ndcg@10": 0.8714056173186473,
|
621 |
+
"precision@1": 0.753,
|
622 |
+
"precision@5": 0.18980000000000002,
|
623 |
+
"precision@10": 0.0981,
|
624 |
+
"recall@1": 0.753,
|
625 |
+
"recall@5": 0.949,
|
626 |
+
"recall@10": 0.981,
|
627 |
+
"f1@1": 0.753,
|
628 |
+
"f1@5": 0.3163333333333333,
|
629 |
+
"f1@10": 0.1783636363636363,
|
630 |
+
"map@1": 0.753,
|
631 |
+
"map@5": 0.8315833333333333,
|
632 |
+
"map@10": 0.83575,
|
633 |
+
"mrr@1": 0.753,
|
634 |
+
"mrr@5": 0.8315833333333333,
|
635 |
+
"mrr@10": 0.83575,
|
636 |
+
"num_pred": 1000,
|
637 |
+
"num_data": 1000
|
638 |
+
},
|
639 |
+
"MSCOCO_i2t": {
|
640 |
+
"hit@1": 0.712,
|
641 |
+
"hit@5": 0.938,
|
642 |
+
"hit@10": 0.983,
|
643 |
+
"ndcg@1": 0.712,
|
644 |
+
"ndcg@5": 0.8365187186277351,
|
645 |
+
"ndcg@10": 0.8515553150974686,
|
646 |
+
"precision@1": 0.712,
|
647 |
+
"precision@5": 0.18760000000000002,
|
648 |
+
"precision@10": 0.09830000000000001,
|
649 |
+
"recall@1": 0.712,
|
650 |
+
"recall@5": 0.938,
|
651 |
+
"recall@10": 0.983,
|
652 |
+
"f1@1": 0.712,
|
653 |
+
"f1@5": 0.3126666666666667,
|
654 |
+
"f1@10": 0.1787272727272727,
|
655 |
+
"map@1": 0.712,
|
656 |
+
"map@5": 0.8023333333333332,
|
657 |
+
"map@10": 0.8088210317460316,
|
658 |
+
"mrr@1": 0.712,
|
659 |
+
"mrr@5": 0.8023333333333332,
|
660 |
+
"mrr@10": 0.8088210317460316,
|
661 |
+
"num_pred": 1000,
|
662 |
+
"num_data": 1000
|
663 |
+
},
|
664 |
+
"NIGHTS": {
|
665 |
+
"hit@1": 0.682,
|
666 |
+
"hit@5": 0.979,
|
667 |
+
"hit@10": 0.996,
|
668 |
+
"ndcg@1": 0.682,
|
669 |
+
"ndcg@5": 0.8523989480099917,
|
670 |
+
"ndcg@10": 0.8581914739405201,
|
671 |
+
"precision@1": 0.682,
|
672 |
+
"precision@5": 0.1958,
|
673 |
+
"precision@10": 0.09960000000000002,
|
674 |
+
"recall@1": 0.682,
|
675 |
+
"recall@5": 0.979,
|
676 |
+
"recall@10": 0.996,
|
677 |
+
"f1@1": 0.682,
|
678 |
+
"f1@5": 0.3263333333333333,
|
679 |
+
"f1@10": 0.18109090909090905,
|
680 |
+
"map@1": 0.682,
|
681 |
+
"map@5": 0.8091166666666667,
|
682 |
+
"map@10": 0.8116809523809525,
|
683 |
+
"mrr@1": 0.682,
|
684 |
+
"mrr@5": 0.8091166666666667,
|
685 |
+
"mrr@10": 0.8116809523809525,
|
686 |
+
"num_pred": 1000,
|
687 |
+
"num_data": 1000
|
688 |
+
},
|
689 |
+
"WebQA": {
|
690 |
+
"hit@1": 0.904,
|
691 |
+
"hit@5": 0.988,
|
692 |
+
"hit@10": 0.993,
|
693 |
+
"ndcg@1": 0.904,
|
694 |
+
"ndcg@5": 0.9525844011004274,
|
695 |
+
"ndcg@10": 0.9543018208718704,
|
696 |
+
"precision@1": 0.904,
|
697 |
+
"precision@5": 0.19760000000000003,
|
698 |
+
"precision@10": 0.09930000000000001,
|
699 |
+
"recall@1": 0.904,
|
700 |
+
"recall@5": 0.988,
|
701 |
+
"recall@10": 0.993,
|
702 |
+
"f1@1": 0.904,
|
703 |
+
"f1@5": 0.3293333333333333,
|
704 |
+
"f1@10": 0.18054545454545454,
|
705 |
+
"map@1": 0.904,
|
706 |
+
"map@5": 0.9404333333333333,
|
707 |
+
"map@10": 0.9412011904761906,
|
708 |
+
"mrr@1": 0.904,
|
709 |
+
"mrr@5": 0.9404333333333333,
|
710 |
+
"mrr@10": 0.9412011904761906,
|
711 |
+
"num_pred": 1000,
|
712 |
+
"num_data": 1000
|
713 |
+
},
|
714 |
+
"FashionIQ": {
|
715 |
+
"hit@1": 0.198,
|
716 |
+
"hit@5": 0.434,
|
717 |
+
"hit@10": 0.576,
|
718 |
+
"ndcg@1": 0.198,
|
719 |
+
"ndcg@5": 0.3202686234787419,
|
720 |
+
"ndcg@10": 0.3663347345246503,
|
721 |
+
"precision@1": 0.198,
|
722 |
+
"precision@5": 0.08680000000000002,
|
723 |
+
"precision@10": 0.05759999999999999,
|
724 |
+
"recall@1": 0.198,
|
725 |
+
"recall@5": 0.434,
|
726 |
+
"recall@10": 0.576,
|
727 |
+
"f1@1": 0.198,
|
728 |
+
"f1@5": 0.1446666666666667,
|
729 |
+
"f1@10": 0.10472727272727275,
|
730 |
+
"map@1": 0.198,
|
731 |
+
"map@5": 0.28276666666666667,
|
732 |
+
"map@10": 0.3018611111111111,
|
733 |
+
"mrr@1": 0.198,
|
734 |
+
"mrr@5": 0.28276666666666667,
|
735 |
+
"mrr@10": 0.3018611111111111,
|
736 |
+
"num_pred": 1000,
|
737 |
+
"num_data": 1000
|
738 |
+
},
|
739 |
+
"Wiki-SS-NQ": {
|
740 |
+
"hit@1": 0.673,
|
741 |
+
"hit@5": 0.88,
|
742 |
+
"hit@10": 0.924,
|
743 |
+
"ndcg@1": 0.673,
|
744 |
+
"ndcg@5": 0.7884163352195437,
|
745 |
+
"ndcg@10": 0.8025271896569116,
|
746 |
+
"precision@1": 0.673,
|
747 |
+
"precision@5": 0.17600000000000002,
|
748 |
+
"precision@10": 0.09240000000000001,
|
749 |
+
"recall@1": 0.673,
|
750 |
+
"recall@5": 0.88,
|
751 |
+
"recall@10": 0.924,
|
752 |
+
"f1@1": 0.673,
|
753 |
+
"f1@5": 0.2933333333333334,
|
754 |
+
"f1@10": 0.168,
|
755 |
+
"map@1": 0.673,
|
756 |
+
"map@5": 0.7574833333333334,
|
757 |
+
"map@10": 0.7632420634920635,
|
758 |
+
"mrr@1": 0.673,
|
759 |
+
"mrr@5": 0.7574833333333334,
|
760 |
+
"mrr@10": 0.7632420634920635,
|
761 |
+
"num_pred": 1000,
|
762 |
+
"num_data": 1000
|
763 |
+
},
|
764 |
+
"OVEN": {
|
765 |
+
"hit@1": 0.643,
|
766 |
+
"hit@5": 0.85,
|
767 |
+
"hit@10": 0.906,
|
768 |
+
"ndcg@1": 0.643,
|
769 |
+
"ndcg@5": 0.7592547199244455,
|
770 |
+
"ndcg@10": 0.7774573209931223,
|
771 |
+
"precision@1": 0.643,
|
772 |
+
"precision@5": 0.17,
|
773 |
+
"precision@10": 0.09060000000000001,
|
774 |
+
"recall@1": 0.643,
|
775 |
+
"recall@5": 0.85,
|
776 |
+
"recall@10": 0.906,
|
777 |
+
"f1@1": 0.643,
|
778 |
+
"f1@5": 0.2833333333333334,
|
779 |
+
"f1@10": 0.16472727272727272,
|
780 |
+
"map@1": 0.643,
|
781 |
+
"map@5": 0.7286,
|
782 |
+
"map@10": 0.7361630952380951,
|
783 |
+
"mrr@1": 0.643,
|
784 |
+
"mrr@5": 0.7286,
|
785 |
+
"mrr@10": 0.7361630952380951,
|
786 |
+
"num_pred": 1000,
|
787 |
+
"num_data": 1000
|
788 |
+
},
|
789 |
+
"EDIS": {
|
790 |
+
"hit@1": 0.843,
|
791 |
+
"hit@5": 0.981,
|
792 |
+
"hit@10": 0.991,
|
793 |
+
"ndcg@1": 0.843,
|
794 |
+
"ndcg@5": 0.9199736805813908,
|
795 |
+
"ndcg@10": 0.923294732165963,
|
796 |
+
"precision@1": 0.843,
|
797 |
+
"precision@5": 0.19620000000000004,
|
798 |
+
"precision@10": 0.09910000000000002,
|
799 |
+
"recall@1": 0.843,
|
800 |
+
"recall@5": 0.981,
|
801 |
+
"recall@10": 0.991,
|
802 |
+
"f1@1": 0.843,
|
803 |
+
"f1@5": 0.327,
|
804 |
+
"f1@10": 0.18018181818181817,
|
805 |
+
"map@1": 0.843,
|
806 |
+
"map@5": 0.8993833333333332,
|
807 |
+
"map@10": 0.9008039682539682,
|
808 |
+
"mrr@1": 0.843,
|
809 |
+
"mrr@5": 0.8993833333333332,
|
810 |
+
"mrr@10": 0.9008039682539682,
|
811 |
+
"num_pred": 1000,
|
812 |
+
"num_data": 1000
|
813 |
+
},
|
814 |
+
"MSCOCO": {
|
815 |
+
"hit@1": 0.668,
|
816 |
+
"hit@5": 0.798,
|
817 |
+
"hit@10": 0.845,
|
818 |
+
"ndcg@1": 0.668,
|
819 |
+
"ndcg@5": 0.7379534961210753,
|
820 |
+
"ndcg@10": 0.7530098472988259,
|
821 |
+
"precision@1": 0.668,
|
822 |
+
"precision@5": 0.15960000000000002,
|
823 |
+
"precision@10": 0.0845,
|
824 |
+
"recall@1": 0.668,
|
825 |
+
"recall@5": 0.798,
|
826 |
+
"recall@10": 0.845,
|
827 |
+
"f1@1": 0.668,
|
828 |
+
"f1@5": 0.266,
|
829 |
+
"f1@10": 0.15363636363636363,
|
830 |
+
"map@1": 0.668,
|
831 |
+
"map@5": 0.7179166666666668,
|
832 |
+
"map@10": 0.7240496031746031,
|
833 |
+
"mrr@1": 0.668,
|
834 |
+
"mrr@5": 0.7179166666666668,
|
835 |
+
"mrr@10": 0.7240496031746031,
|
836 |
+
"num_pred": 1000,
|
837 |
+
"num_data": 1000
|
838 |
+
},
|
839 |
+
"RefCOCO": {
|
840 |
+
"hit@1": 0.869,
|
841 |
+
"hit@5": 0.977,
|
842 |
+
"hit@10": 0.993,
|
843 |
+
"ndcg@1": 0.869,
|
844 |
+
"ndcg@5": 0.9312989016276231,
|
845 |
+
"ndcg@10": 0.9364920604671089,
|
846 |
+
"precision@1": 0.869,
|
847 |
+
"precision@5": 0.19540000000000005,
|
848 |
+
"precision@10": 0.09930000000000001,
|
849 |
+
"recall@1": 0.869,
|
850 |
+
"recall@5": 0.977,
|
851 |
+
"recall@10": 0.993,
|
852 |
+
"f1@1": 0.869,
|
853 |
+
"f1@5": 0.3256666666666666,
|
854 |
+
"f1@10": 0.18054545454545454,
|
855 |
+
"map@1": 0.869,
|
856 |
+
"map@5": 0.9156666666666666,
|
857 |
+
"map@10": 0.9178246031746032,
|
858 |
+
"mrr@1": 0.869,
|
859 |
+
"mrr@5": 0.9156666666666666,
|
860 |
+
"mrr@10": 0.9178246031746032,
|
861 |
+
"num_pred": 1000,
|
862 |
+
"num_data": 1000
|
863 |
+
},
|
864 |
+
"RefCOCO-Matching": {
|
865 |
+
"hit@1": 0.859,
|
866 |
+
"hit@5": 1.0,
|
867 |
+
"hit@10": 1.0,
|
868 |
+
"ndcg@1": 0.859,
|
869 |
+
"ndcg@5": 0.9490167651117406,
|
870 |
+
"ndcg@10": 0.9490167651117406,
|
871 |
+
"precision@1": 0.859,
|
872 |
+
"precision@5": 0.20000000000000004,
|
873 |
+
"precision@10": 0.10000000000000002,
|
874 |
+
"recall@1": 0.859,
|
875 |
+
"recall@5": 1.0,
|
876 |
+
"recall@10": 1.0,
|
877 |
+
"f1@1": 0.859,
|
878 |
+
"f1@5": 0.3333333333333333,
|
879 |
+
"f1@10": 0.18181818181818182,
|
880 |
+
"map@1": 0.859,
|
881 |
+
"map@5": 0.93145,
|
882 |
+
"map@10": 0.93145,
|
883 |
+
"mrr@1": 0.859,
|
884 |
+
"mrr@5": 0.92845,
|
885 |
+
"mrr@10": 0.92845,
|
886 |
+
"num_pred": 1000,
|
887 |
+
"num_data": 1000
|
888 |
+
},
|
889 |
+
"Visual7W-Pointing": {
|
890 |
+
"hit@1": 0.69,
|
891 |
+
"hit@5": 0.865,
|
892 |
+
"hit@10": 0.906,
|
893 |
+
"ndcg@1": 0.69,
|
894 |
+
"ndcg@5": 0.78750158149685,
|
895 |
+
"ndcg@10": 0.8006061099313997,
|
896 |
+
"precision@1": 0.69,
|
897 |
+
"precision@5": 0.173,
|
898 |
+
"precision@10": 0.09060000000000001,
|
899 |
+
"recall@1": 0.69,
|
900 |
+
"recall@5": 0.865,
|
901 |
+
"recall@10": 0.906,
|
902 |
+
"f1@1": 0.69,
|
903 |
+
"f1@5": 0.2883333333333334,
|
904 |
+
"f1@10": 0.16472727272727272,
|
905 |
+
"map@1": 0.69,
|
906 |
+
"map@5": 0.7613333333333334,
|
907 |
+
"map@10": 0.7666503968253968,
|
908 |
+
"mrr@1": 0.69,
|
909 |
+
"mrr@5": 0.7613333333333334,
|
910 |
+
"mrr@10": 0.7666503968253968,
|
911 |
+
"num_pred": 1000,
|
912 |
+
"num_data": 1000
|
913 |
+
}
|
914 |
+
},
|
915 |
+
"visdoc": {
|
916 |
+
"ViDoRe_arxivqa": {
|
917 |
+
"hit@1": 0.73,
|
918 |
+
"hit@5": 0.862,
|
919 |
+
"hit@10": 0.902,
|
920 |
+
"ndcg@1": 0.73,
|
921 |
+
"ndcg@5": 0.8029233527536316,
|
922 |
+
"ndcg@10": 0.8163870890886299,
|
923 |
+
"precision@1": 0.73,
|
924 |
+
"precision@5": 0.17239999999999997,
|
925 |
+
"precision@10": 0.0902,
|
926 |
+
"recall@1": 0.73,
|
927 |
+
"recall@5": 0.862,
|
928 |
+
"recall@10": 0.902,
|
929 |
+
"f1@1": 0.73,
|
930 |
+
"f1@5": 0.2873333333333334,
|
931 |
+
"f1@10": 0.164,
|
932 |
+
"map@1": 0.73,
|
933 |
+
"map@5": 0.7830333333333332,
|
934 |
+
"map@10": 0.7889071428571428,
|
935 |
+
"mrr@1": 0.73,
|
936 |
+
"mrr@5": 0.7830333333333332,
|
937 |
+
"mrr@10": 0.7889071428571428,
|
938 |
+
"num_pred": 500,
|
939 |
+
"num_data": 500
|
940 |
+
},
|
941 |
+
"ViDoRe_docvqa": {
|
942 |
+
"hit@1": 0.34811529933481156,
|
943 |
+
"hit@5": 0.541019955654102,
|
944 |
+
"hit@10": 0.6208425720620843,
|
945 |
+
"ndcg@1": 0.34811529933481156,
|
946 |
+
"ndcg@5": 0.447298130737004,
|
947 |
+
"ndcg@10": 0.472700870616564,
|
948 |
+
"precision@1": 0.34811529933481156,
|
949 |
+
"precision@5": 0.10909090909090907,
|
950 |
+
"precision@10": 0.06341463414634145,
|
951 |
+
"recall@1": 0.3409830007390983,
|
952 |
+
"recall@5": 0.5324833702882483,
|
953 |
+
"recall@10": 0.6106060606060606,
|
954 |
+
"f1@1": 0.342543635226562,
|
955 |
+
"f1@5": 0.17995677206581082,
|
956 |
+
"f1@10": 0.11417931661834103,
|
957 |
+
"map@1": 0.34811529933481156,
|
958 |
+
"map@5": 0.4164326188716433,
|
959 |
+
"map@10": 0.42694594023862315,
|
960 |
+
"mrr@1": 0.34811529933481156,
|
961 |
+
"mrr@5": 0.4236881005173688,
|
962 |
+
"mrr@10": 0.4339888431351846,
|
963 |
+
"num_pred": 451,
|
964 |
+
"num_data": 451
|
965 |
+
},
|
966 |
+
"ViDoRe_infovqa": {
|
967 |
+
"hit@1": 0.7753036437246964,
|
968 |
+
"hit@5": 0.8967611336032388,
|
969 |
+
"hit@10": 0.9433198380566802,
|
970 |
+
"ndcg@1": 0.7753036437246964,
|
971 |
+
"ndcg@5": 0.8375656543815698,
|
972 |
+
"ndcg@10": 0.8530816084621412,
|
973 |
+
"precision@1": 0.7753036437246964,
|
974 |
+
"precision@5": 0.17935222672064777,
|
975 |
+
"precision@10": 0.0945344129554656,
|
976 |
+
"recall@1": 0.7753036437246964,
|
977 |
+
"recall@5": 0.895748987854251,
|
978 |
+
"recall@10": 0.9433198380566802,
|
979 |
+
"f1@1": 0.7753036437246964,
|
980 |
+
"f1@5": 0.29882398303450936,
|
981 |
+
"f1@10": 0.1718194086615139,
|
982 |
+
"map@1": 0.7753036437246964,
|
983 |
+
"map@5": 0.8181848852901483,
|
984 |
+
"map@10": 0.8247228648544437,
|
985 |
+
"mrr@1": 0.7753036437246964,
|
986 |
+
"mrr@5": 0.8185222672064777,
|
987 |
+
"mrr@10": 0.8247710622710623,
|
988 |
+
"num_pred": 494,
|
989 |
+
"num_data": 494
|
990 |
+
},
|
991 |
+
"ViDoRe_tabfquad": {
|
992 |
+
"hit@1": 0.8107142857142857,
|
993 |
+
"hit@5": 0.9571428571428572,
|
994 |
+
"hit@10": 0.975,
|
995 |
+
"ndcg@1": 0.8107142857142857,
|
996 |
+
"ndcg@5": 0.8920767921191974,
|
997 |
+
"ndcg@10": 0.8982104341600656,
|
998 |
+
"precision@1": 0.8107142857142857,
|
999 |
+
"precision@5": 0.19142857142857142,
|
1000 |
+
"precision@10": 0.09749999999999999,
|
1001 |
+
"recall@1": 0.8107142857142857,
|
1002 |
+
"recall@5": 0.9571428571428572,
|
1003 |
+
"recall@10": 0.975,
|
1004 |
+
"f1@1": 0.8107142857142857,
|
1005 |
+
"f1@5": 0.31904761904761914,
|
1006 |
+
"f1@10": 0.1772727272727273,
|
1007 |
+
"map@1": 0.8107142857142857,
|
1008 |
+
"map@5": 0.8701190476190476,
|
1009 |
+
"map@10": 0.8728613945578231,
|
1010 |
+
"mrr@1": 0.8107142857142857,
|
1011 |
+
"mrr@5": 0.8701190476190476,
|
1012 |
+
"mrr@10": 0.8728613945578231,
|
1013 |
+
"num_pred": 280,
|
1014 |
+
"num_data": 280
|
1015 |
+
},
|
1016 |
+
"ViDoRe_tatdqa": {
|
1017 |
+
"hit@1": 0.30984204131227217,
|
1018 |
+
"hit@5": 0.554678007290401,
|
1019 |
+
"hit@10": 0.6458080194410692,
|
1020 |
+
"ndcg@1": 0.30984204131227217,
|
1021 |
+
"ndcg@5": 0.4379457147354835,
|
1022 |
+
"ndcg@10": 0.4670932550977435,
|
1023 |
+
"precision@1": 0.30984204131227217,
|
1024 |
+
"precision@5": 0.11117861482381532,
|
1025 |
+
"precision@10": 0.06476306196840825,
|
1026 |
+
"recall@1": 0.30893074119076547,
|
1027 |
+
"recall@5": 0.5540704738760632,
|
1028 |
+
"recall@10": 0.6446537059538274,
|
1029 |
+
"f1@1": 0.3092345078979344,
|
1030 |
+
"f1@5": 0.18512411039750049,
|
1031 |
+
"f1@10": 0.11762767406752828,
|
1032 |
+
"map@1": 0.30984204131227217,
|
1033 |
+
"map@5": 0.3993114621304172,
|
1034 |
+
"map@10": 0.4112191170514378,
|
1035 |
+
"mrr@1": 0.30984204131227217,
|
1036 |
+
"mrr@5": 0.3997772377480761,
|
1037 |
+
"mrr@10": 0.4117528785511775,
|
1038 |
+
"num_pred": 1646,
|
1039 |
+
"num_data": 1646
|
1040 |
+
},
|
1041 |
+
"ViDoRe_shiftproject": {
|
1042 |
+
"hit@1": 0.43,
|
1043 |
+
"hit@5": 0.72,
|
1044 |
+
"hit@10": 0.86,
|
1045 |
+
"ndcg@1": 0.43,
|
1046 |
+
"ndcg@5": 0.5929154109940404,
|
1047 |
+
"ndcg@10": 0.6392206165229328,
|
1048 |
+
"precision@1": 0.43,
|
1049 |
+
"precision@5": 0.14399999999999996,
|
1050 |
+
"precision@10": 0.08599999999999998,
|
1051 |
+
"recall@1": 0.43,
|
1052 |
+
"recall@5": 0.72,
|
1053 |
+
"recall@10": 0.86,
|
1054 |
+
"f1@1": 0.43,
|
1055 |
+
"f1@5": 0.24,
|
1056 |
+
"f1@10": 0.15636363636363637,
|
1057 |
+
"map@1": 0.43,
|
1058 |
+
"map@5": 0.5498333333333334,
|
1059 |
+
"map@10": 0.5695357142857143,
|
1060 |
+
"mrr@1": 0.43,
|
1061 |
+
"mrr@5": 0.5498333333333334,
|
1062 |
+
"mrr@10": 0.5695357142857143,
|
1063 |
+
"num_pred": 100,
|
1064 |
+
"num_data": 100
|
1065 |
+
},
|
1066 |
+
"ViDoRe_syntheticDocQA_artificial_intelligence": {
|
1067 |
+
"hit@1": 0.76,
|
1068 |
+
"hit@5": 0.97,
|
1069 |
+
"hit@10": 0.98,
|
1070 |
+
"ndcg@1": 0.76,
|
1071 |
+
"ndcg@5": 0.8826369310807379,
|
1072 |
+
"ndcg@10": 0.8857915798485954,
|
1073 |
+
"precision@1": 0.76,
|
1074 |
+
"precision@5": 0.19399999999999995,
|
1075 |
+
"precision@10": 0.09799999999999998,
|
1076 |
+
"recall@1": 0.76,
|
1077 |
+
"recall@5": 0.97,
|
1078 |
+
"recall@10": 0.98,
|
1079 |
+
"f1@1": 0.76,
|
1080 |
+
"f1@5": 0.3233333333333334,
|
1081 |
+
"f1@10": 0.17818181818181827,
|
1082 |
+
"map@1": 0.76,
|
1083 |
+
"map@5": 0.8524999999999998,
|
1084 |
+
"map@10": 0.8537499999999999,
|
1085 |
+
"mrr@1": 0.76,
|
1086 |
+
"mrr@5": 0.8524999999999998,
|
1087 |
+
"mrr@10": 0.8537499999999999,
|
1088 |
+
"num_pred": 100,
|
1089 |
+
"num_data": 100
|
1090 |
+
},
|
1091 |
+
"ViDoRe_syntheticDocQA_energy": {
|
1092 |
+
"hit@1": 0.79,
|
1093 |
+
"hit@5": 0.92,
|
1094 |
+
"hit@10": 0.97,
|
1095 |
+
"ndcg@1": 0.79,
|
1096 |
+
"ndcg@5": 0.8607760819564892,
|
1097 |
+
"ndcg@10": 0.8773985077564801,
|
1098 |
+
"precision@1": 0.79,
|
1099 |
+
"precision@5": 0.18399999999999994,
|
1100 |
+
"precision@10": 0.09699999999999999,
|
1101 |
+
"recall@1": 0.79,
|
1102 |
+
"recall@5": 0.92,
|
1103 |
+
"recall@10": 0.97,
|
1104 |
+
"f1@1": 0.79,
|
1105 |
+
"f1@5": 0.3066666666666667,
|
1106 |
+
"f1@10": 0.17636363636363647,
|
1107 |
+
"map@1": 0.79,
|
1108 |
+
"map@5": 0.8408333333333333,
|
1109 |
+
"map@10": 0.8479563492063491,
|
1110 |
+
"mrr@1": 0.79,
|
1111 |
+
"mrr@5": 0.8408333333333333,
|
1112 |
+
"mrr@10": 0.8479563492063491,
|
1113 |
+
"num_pred": 100,
|
1114 |
+
"num_data": 100
|
1115 |
+
},
|
1116 |
+
"ViDoRe_syntheticDocQA_government_reports": {
|
1117 |
+
"hit@1": 0.75,
|
1118 |
+
"hit@5": 0.94,
|
1119 |
+
"hit@10": 0.98,
|
1120 |
+
"ndcg@1": 0.75,
|
1121 |
+
"ndcg@5": 0.8535725026359785,
|
1122 |
+
"ndcg@10": 0.8664541719994131,
|
1123 |
+
"precision@1": 0.75,
|
1124 |
+
"precision@5": 0.18799999999999997,
|
1125 |
+
"precision@10": 0.09799999999999999,
|
1126 |
+
"recall@1": 0.75,
|
1127 |
+
"recall@5": 0.94,
|
1128 |
+
"recall@10": 0.98,
|
1129 |
+
"f1@1": 0.75,
|
1130 |
+
"f1@5": 0.3133333333333334,
|
1131 |
+
"f1@10": 0.17818181818181825,
|
1132 |
+
"map@1": 0.75,
|
1133 |
+
"map@5": 0.8245,
|
1134 |
+
"map@10": 0.8297777777777777,
|
1135 |
+
"mrr@1": 0.75,
|
1136 |
+
"mrr@5": 0.8245,
|
1137 |
+
"mrr@10": 0.8297777777777777,
|
1138 |
+
"num_pred": 100,
|
1139 |
+
"num_data": 100
|
1140 |
+
},
|
1141 |
+
"ViDoRe_syntheticDocQA_healthcare_industry": {
|
1142 |
+
"hit@1": 0.82,
|
1143 |
+
"hit@5": 0.98,
|
1144 |
+
"hit@10": 0.99,
|
1145 |
+
"ndcg@1": 0.82,
|
1146 |
+
"ndcg@5": 0.9150183360093088,
|
1147 |
+
"ndcg@10": 0.9185804078803891,
|
1148 |
+
"precision@1": 0.82,
|
1149 |
+
"precision@5": 0.19599999999999998,
|
1150 |
+
"precision@10": 0.09899999999999999,
|
1151 |
+
"recall@1": 0.82,
|
1152 |
+
"recall@5": 0.98,
|
1153 |
+
"recall@10": 0.99,
|
1154 |
+
"f1@1": 0.82,
|
1155 |
+
"f1@5": 0.32666666666666677,
|
1156 |
+
"f1@10": 0.18000000000000008,
|
1157 |
+
"map@1": 0.82,
|
1158 |
+
"map@5": 0.8925,
|
1159 |
+
"map@10": 0.8941666666666667,
|
1160 |
+
"mrr@1": 0.82,
|
1161 |
+
"mrr@5": 0.8925,
|
1162 |
+
"mrr@10": 0.8941666666666667,
|
1163 |
+
"num_pred": 100,
|
1164 |
+
"num_data": 100
|
1165 |
+
},
|
1166 |
+
"VisRAG_ArxivQA": {
|
1167 |
+
"hit@1": 0.6838235294117647,
|
1168 |
+
"hit@5": 0.8125,
|
1169 |
+
"hit@10": 0.8627450980392157,
|
1170 |
+
"ndcg@1": 0.6838235294117647,
|
1171 |
+
"ndcg@5": 0.7543475434444639,
|
1172 |
+
"ndcg@10": 0.7706335822507916,
|
1173 |
+
"precision@1": 0.6838235294117647,
|
1174 |
+
"precision@5": 0.16249999999999995,
|
1175 |
+
"precision@10": 0.08627450980392158,
|
1176 |
+
"recall@1": 0.6838235294117647,
|
1177 |
+
"recall@5": 0.8125,
|
1178 |
+
"recall@10": 0.8627450980392157,
|
1179 |
+
"f1@1": 0.6838235294117647,
|
1180 |
+
"f1@5": 0.2708333333333334,
|
1181 |
+
"f1@10": 0.1568627450980392,
|
1182 |
+
"map@1": 0.6838235294117647,
|
1183 |
+
"map@5": 0.7348243464052288,
|
1184 |
+
"map@10": 0.741572323373794,
|
1185 |
+
"mrr@1": 0.6838235294117647,
|
1186 |
+
"mrr@5": 0.7348243464052288,
|
1187 |
+
"mrr@10": 0.741572323373794,
|
1188 |
+
"num_pred": 816,
|
1189 |
+
"num_data": 816
|
1190 |
+
},
|
1191 |
+
"VisRAG_ChartQA": {
|
1192 |
+
"hit@1": 0.7301587301587301,
|
1193 |
+
"hit@5": 0.9047619047619048,
|
1194 |
+
"hit@10": 0.9365079365079365,
|
1195 |
+
"ndcg@1": 0.7301587301587301,
|
1196 |
+
"ndcg@5": 0.8291117085763197,
|
1197 |
+
"ndcg@10": 0.8386682163751762,
|
1198 |
+
"precision@1": 0.7301587301587301,
|
1199 |
+
"precision@5": 0.18095238095238086,
|
1200 |
+
"precision@10": 0.0936507936507936,
|
1201 |
+
"recall@1": 0.7301587301587301,
|
1202 |
+
"recall@5": 0.9047619047619048,
|
1203 |
+
"recall@10": 0.9365079365079365,
|
1204 |
+
"f1@1": 0.7301587301587301,
|
1205 |
+
"f1@5": 0.3015873015873015,
|
1206 |
+
"f1@10": 0.1702741702741703,
|
1207 |
+
"map@1": 0.7301587301587301,
|
1208 |
+
"map@5": 0.8034391534391533,
|
1209 |
+
"map@10": 0.8069664902998236,
|
1210 |
+
"mrr@1": 0.7301587301587301,
|
1211 |
+
"mrr@5": 0.8034391534391533,
|
1212 |
+
"mrr@10": 0.8069664902998236,
|
1213 |
+
"num_pred": 63,
|
1214 |
+
"num_data": 63
|
1215 |
+
},
|
1216 |
+
"VisRAG_MP-DocVQA": {
|
1217 |
+
"hit@1": 0.622673434856176,
|
1218 |
+
"hit@5": 0.8274111675126904,
|
1219 |
+
"hit@10": 0.8697123519458545,
|
1220 |
+
"ndcg@1": 0.622673434856176,
|
1221 |
+
"ndcg@5": 0.7387055085318691,
|
1222 |
+
"ndcg@10": 0.7527120281213929,
|
1223 |
+
"precision@1": 0.622673434856176,
|
1224 |
+
"precision@5": 0.16548223350253805,
|
1225 |
+
"precision@10": 0.08697123519458544,
|
1226 |
+
"recall@1": 0.622673434856176,
|
1227 |
+
"recall@5": 0.8274111675126904,
|
1228 |
+
"recall@10": 0.8697123519458545,
|
1229 |
+
"f1@1": 0.622673434856176,
|
1230 |
+
"f1@5": 0.27580372250423024,
|
1231 |
+
"f1@10": 0.15812951853560991,
|
1232 |
+
"map@1": 0.622673434856176,
|
1233 |
+
"map@5": 0.7085730400451212,
|
1234 |
+
"map@10": 0.7145482502081486,
|
1235 |
+
"mrr@1": 0.622673434856176,
|
1236 |
+
"mrr@5": 0.7085730400451212,
|
1237 |
+
"mrr@10": 0.7145482502081486,
|
1238 |
+
"num_pred": 591,
|
1239 |
+
"num_data": 591
|
1240 |
+
},
|
1241 |
+
"VisRAG_SlideVQA": {
|
1242 |
+
"hit@1": 0.8794964028776978,
|
1243 |
+
"hit@5": 0.9784172661870504,
|
1244 |
+
"hit@10": 0.9856115107913669,
|
1245 |
+
"ndcg@1": 0.8794964028776978,
|
1246 |
+
"ndcg@5": 0.9188068852639469,
|
1247 |
+
"ndcg@10": 0.9244602518626103,
|
1248 |
+
"precision@1": 0.8794964028776978,
|
1249 |
+
"precision@5": 0.2399280575539568,
|
1250 |
+
"precision@10": 0.1223021582733813,
|
1251 |
+
"recall@1": 0.7634892086330936,
|
1252 |
+
"recall@5": 0.9595323741007195,
|
1253 |
+
"recall@10": 0.9748201438848921,
|
1254 |
+
"f1@1": 0.802158273381295,
|
1255 |
+
"f1@5": 0.3770126755738267,
|
1256 |
+
"f1@10": 0.21484630477436234,
|
1257 |
+
"map@1": 0.8794964028776978,
|
1258 |
+
"map@5": 0.8942146282973621,
|
1259 |
+
"map@10": 0.8974313406417723,
|
1260 |
+
"mrr@1": 0.8794964028776978,
|
1261 |
+
"mrr@5": 0.9245203836930455,
|
1262 |
+
"mrr@10": 0.9256337786913328,
|
1263 |
+
"num_pred": 556,
|
1264 |
+
"num_data": 556
|
1265 |
+
},
|
1266 |
+
"VisRAG_InfoVQA": {
|
1267 |
+
"hit@1": 0.7562674094707521,
|
1268 |
+
"hit@5": 0.9373259052924791,
|
1269 |
+
"hit@10": 0.9596100278551533,
|
1270 |
+
"ndcg@1": 0.7562674094707521,
|
1271 |
+
"ndcg@5": 0.8598286164145738,
|
1272 |
+
"ndcg@10": 0.8670879150249058,
|
1273 |
+
"precision@1": 0.7562674094707521,
|
1274 |
+
"precision@5": 0.18746518105849577,
|
1275 |
+
"precision@10": 0.0959610027855153,
|
1276 |
+
"recall@1": 0.7562674094707521,
|
1277 |
+
"recall@5": 0.9373259052924791,
|
1278 |
+
"recall@10": 0.9596100278551533,
|
1279 |
+
"f1@1": 0.7562674094707521,
|
1280 |
+
"f1@5": 0.31244196843082644,
|
1281 |
+
"f1@10": 0.1744745505191188,
|
1282 |
+
"map@1": 0.7562674094707521,
|
1283 |
+
"map@5": 0.8333565459610028,
|
1284 |
+
"map@10": 0.8363808197373656,
|
1285 |
+
"mrr@1": 0.7562674094707521,
|
1286 |
+
"mrr@5": 0.8333565459610028,
|
1287 |
+
"mrr@10": 0.8363808197373656,
|
1288 |
+
"num_pred": 718,
|
1289 |
+
"num_data": 718
|
1290 |
+
},
|
1291 |
+
"VisRAG_PlotQA": {
|
1292 |
+
"hit@1": 0.507531865585168,
|
1293 |
+
"hit@5": 0.7879490150637312,
|
1294 |
+
"hit@10": 0.8887601390498262,
|
1295 |
+
"ndcg@1": 0.507531865585168,
|
1296 |
+
"ndcg@5": 0.6575765009394535,
|
1297 |
+
"ndcg@10": 0.6903891309073118,
|
1298 |
+
"precision@1": 0.507531865585168,
|
1299 |
+
"precision@5": 0.15758980301274622,
|
1300 |
+
"precision@10": 0.08887601390498262,
|
1301 |
+
"recall@1": 0.507531865585168,
|
1302 |
+
"recall@5": 0.7879490150637312,
|
1303 |
+
"recall@10": 0.8887601390498262,
|
1304 |
+
"f1@1": 0.507531865585168,
|
1305 |
+
"f1@5": 0.2626496716879105,
|
1306 |
+
"f1@10": 0.1615927525545139,
|
1307 |
+
"map@1": 0.507531865585168,
|
1308 |
+
"map@5": 0.614156044804944,
|
1309 |
+
"map@10": 0.6278182420129118,
|
1310 |
+
"mrr@1": 0.507531865585168,
|
1311 |
+
"mrr@5": 0.614156044804944,
|
1312 |
+
"mrr@10": 0.6278182420129118,
|
1313 |
+
"num_pred": 863,
|
1314 |
+
"num_data": 863
|
1315 |
+
},
|
1316 |
+
"ViDoSeek-page": {
|
1317 |
+
"hit@1": 0.03502626970227671,
|
1318 |
+
"hit@5": 0.39842381786339753,
|
1319 |
+
"hit@10": 0.5683012259194395,
|
1320 |
+
"ndcg@1": 0.03502626970227671,
|
1321 |
+
"ndcg@5": 0.21966337171000339,
|
1322 |
+
"ndcg@10": 0.2753996989023275,
|
1323 |
+
"precision@1": 0.03502626970227671,
|
1324 |
+
"precision@5": 0.0796847635726795,
|
1325 |
+
"precision@10": 0.05683012259194396,
|
1326 |
+
"recall@1": 0.03502626970227671,
|
1327 |
+
"recall@5": 0.39842381786339753,
|
1328 |
+
"recall@10": 0.5683012259194395,
|
1329 |
+
"f1@1": 0.03502626970227671,
|
1330 |
+
"f1@5": 0.13280793928779921,
|
1331 |
+
"f1@10": 0.10332749562171628,
|
1332 |
+
"map@1": 0.03502626970227671,
|
1333 |
+
"map@5": 0.16103327495621717,
|
1334 |
+
"map@10": 0.1845019181052456,
|
1335 |
+
"mrr@1": 0.03502626970227671,
|
1336 |
+
"mrr@5": 0.16103327495621717,
|
1337 |
+
"mrr@10": 0.1845019181052456,
|
1338 |
+
"num_pred": 1142,
|
1339 |
+
"num_data": 1142
|
1340 |
+
},
|
1341 |
+
"ViDoSeek-doc": {
|
1342 |
+
"hit@1": 0.9903677758318739,
|
1343 |
+
"hit@5": 0.9973730297723292,
|
1344 |
+
"hit@10": 0.9982486865148862,
|
1345 |
+
"ndcg@1": 0.9903677758318739,
|
1346 |
+
"ndcg@5": 0.9017374324785984,
|
1347 |
+
"ndcg@10": 0.8219510951135176,
|
1348 |
+
"precision@1": 0.9903677758318739,
|
1349 |
+
"precision@5": 0.874430823117338,
|
1350 |
+
"precision@10": 0.7652364273204904,
|
1351 |
+
"recall@1": 0.05845834634908238,
|
1352 |
+
"recall@5": 0.2559779257615508,
|
1353 |
+
"recall@10": 0.43857251867789215,
|
1354 |
+
"f1@1": 0.1096528022965413,
|
1355 |
+
"f1@5": 0.38812368632589883,
|
1356 |
+
"f1@10": 0.5452152563395831,
|
1357 |
+
"map@1": 0.9903677758318739,
|
1358 |
+
"map@5": 0.8630998248686514,
|
1359 |
+
"map@10": 0.7442025955643982,
|
1360 |
+
"mrr@1": 0.9903677758318739,
|
1361 |
+
"mrr@5": 0.9936077057793345,
|
1362 |
+
"mrr@10": 0.9937536485697606,
|
1363 |
+
"num_pred": 1142,
|
1364 |
+
"num_data": 1142
|
1365 |
+
},
|
1366 |
+
"MMLongBench-page": {
|
1367 |
+
"hit@1": 0.050119331742243436,
|
1368 |
+
"hit@5": 0.24343675417661098,
|
1369 |
+
"hit@10": 0.34009546539379476,
|
1370 |
+
"ndcg@1": 0.050119331742243436,
|
1371 |
+
"ndcg@5": 0.11926870978586242,
|
1372 |
+
"ndcg@10": 0.15036346679845933,
|
1373 |
+
"precision@1": 0.050119331742243436,
|
1374 |
+
"precision@5": 0.05704057279236276,
|
1375 |
+
"precision@10": 0.04343675417661098,
|
1376 |
+
"recall@1": 0.027645186953062846,
|
1377 |
+
"recall@5": 0.17770627344016365,
|
1378 |
+
"recall@10": 0.26356404136833733,
|
1379 |
+
"f1@1": 0.03427268832996756,
|
1380 |
+
"f1@5": 0.081668369068341,
|
1381 |
+
"f1@10": 0.07102416353802142,
|
1382 |
+
"map@1": 0.050119331742243436,
|
1383 |
+
"map@5": 0.08614856801909307,
|
1384 |
+
"map@10": 0.09990828740008333,
|
1385 |
+
"mrr@1": 0.050119331742243436,
|
1386 |
+
"mrr@5": 0.12074383452665076,
|
1387 |
+
"mrr@10": 0.13376614766829564,
|
1388 |
+
"num_pred": 838,
|
1389 |
+
"num_data": 838
|
1390 |
+
},
|
1391 |
+
"MMLongBench-doc": {
|
1392 |
+
"hit@1": 0.6121718377088305,
|
1393 |
+
"hit@5": 0.7494033412887828,
|
1394 |
+
"hit@10": 0.7971360381861575,
|
1395 |
+
"ndcg@1": 0.6121718377088305,
|
1396 |
+
"ndcg@5": 0.5085425654244546,
|
1397 |
+
"ndcg@10": 0.4531636698619904,
|
1398 |
+
"precision@1": 0.6121718377088305,
|
1399 |
+
"precision@5": 0.47971360381861583,
|
1400 |
+
"precision@10": 0.4142004773269689,
|
1401 |
+
"recall@1": 0.019539454725697798,
|
1402 |
+
"recall@5": 0.07494267826645917,
|
1403 |
+
"recall@10": 0.125038805779916,
|
1404 |
+
"f1@1": 0.037500929446937684,
|
1405 |
+
"f1@5": 0.12468748938491771,
|
1406 |
+
"f1@10": 0.18079259638973433,
|
1407 |
+
"map@1": 0.6121718377088305,
|
1408 |
+
"map@5": 0.44090692124105013,
|
1409 |
+
"map@10": 0.35958542742651733,
|
1410 |
+
"mrr@1": 0.6121718377088305,
|
1411 |
+
"mrr@5": 0.6670445505171042,
|
1412 |
+
"mrr@10": 0.6731584081524417,
|
1413 |
+
"num_pred": 838,
|
1414 |
+
"num_data": 838
|
1415 |
+
}
|
1416 |
+
},
|
1417 |
+
"video": {
|
1418 |
+
"DiDeMo": {
|
1419 |
+
"hit@1": 0.301792828685259,
|
1420 |
+
"hit@5": 0.5378486055776892,
|
1421 |
+
"hit@10": 0.6314741035856574,
|
1422 |
+
"ndcg@1": 0.301792828685259,
|
1423 |
+
"ndcg@5": 0.42577172689527676,
|
1424 |
+
"ndcg@10": 0.45597396154157394,
|
1425 |
+
"precision@1": 0.301792828685259,
|
1426 |
+
"precision@5": 0.10756972111553784,
|
1427 |
+
"precision@10": 0.06314741035856573,
|
1428 |
+
"recall@1": 0.301792828685259,
|
1429 |
+
"recall@5": 0.5378486055776892,
|
1430 |
+
"recall@10": 0.6314741035856574,
|
1431 |
+
"f1@1": 0.301792828685259,
|
1432 |
+
"f1@5": 0.17928286852589648,
|
1433 |
+
"f1@10": 0.11481347337921044,
|
1434 |
+
"map@1": 0.301792828685259,
|
1435 |
+
"map@5": 0.3886122177954847,
|
1436 |
+
"map@10": 0.40103277366723583,
|
1437 |
+
"mrr@1": 0.301792828685259,
|
1438 |
+
"mrr@5": 0.3886122177954847,
|
1439 |
+
"mrr@10": 0.40103277366723583,
|
1440 |
+
"num_pred": 1004,
|
1441 |
+
"num_data": 1004
|
1442 |
+
},
|
1443 |
+
"MSR-VTT": {
|
1444 |
+
"hit@1": 0.278,
|
1445 |
+
"hit@5": 0.502,
|
1446 |
+
"hit@10": 0.602,
|
1447 |
+
"ndcg@1": 0.278,
|
1448 |
+
"ndcg@5": 0.3947140366184671,
|
1449 |
+
"ndcg@10": 0.42686954179538383,
|
1450 |
+
"precision@1": 0.278,
|
1451 |
+
"precision@5": 0.10039999999999999,
|
1452 |
+
"precision@10": 0.0602,
|
1453 |
+
"recall@1": 0.278,
|
1454 |
+
"recall@5": 0.502,
|
1455 |
+
"recall@10": 0.602,
|
1456 |
+
"f1@1": 0.278,
|
1457 |
+
"f1@5": 0.16733333333333336,
|
1458 |
+
"f1@10": 0.10945454545454547,
|
1459 |
+
"map@1": 0.278,
|
1460 |
+
"map@5": 0.3592833333333333,
|
1461 |
+
"map@10": 0.37245674603174606,
|
1462 |
+
"mrr@1": 0.278,
|
1463 |
+
"mrr@5": 0.3592833333333333,
|
1464 |
+
"mrr@10": 0.37245674603174606,
|
1465 |
+
"num_pred": 1000,
|
1466 |
+
"num_data": 1000
|
1467 |
+
},
|
1468 |
+
"MSVD": {
|
1469 |
+
"hit@1": 0.48208955223880595,
|
1470 |
+
"hit@5": 0.7611940298507462,
|
1471 |
+
"hit@10": 0.844776119402985,
|
1472 |
+
"ndcg@1": 0.48208955223880595,
|
1473 |
+
"ndcg@5": 0.6315868198763833,
|
1474 |
+
"ndcg@10": 0.6587947769114488,
|
1475 |
+
"precision@1": 0.48208955223880595,
|
1476 |
+
"precision@5": 0.15223880597014924,
|
1477 |
+
"precision@10": 0.0844776119402985,
|
1478 |
+
"recall@1": 0.48208955223880595,
|
1479 |
+
"recall@5": 0.7611940298507462,
|
1480 |
+
"recall@10": 0.844776119402985,
|
1481 |
+
"f1@1": 0.48208955223880595,
|
1482 |
+
"f1@5": 0.25373134328358216,
|
1483 |
+
"f1@10": 0.15359565807327005,
|
1484 |
+
"map@1": 0.48208955223880595,
|
1485 |
+
"map@5": 0.5883084577114427,
|
1486 |
+
"map@10": 0.5996475953565507,
|
1487 |
+
"mrr@1": 0.48208955223880595,
|
1488 |
+
"mrr@5": 0.5883084577114427,
|
1489 |
+
"mrr@10": 0.5996475953565507,
|
1490 |
+
"num_pred": 670,
|
1491 |
+
"num_data": 670
|
1492 |
+
},
|
1493 |
+
"VATEX": {
|
1494 |
+
"hit@1": 0.26284055381866905,
|
1495 |
+
"hit@5": 0.502009825815096,
|
1496 |
+
"hit@10": 0.6058508262617239,
|
1497 |
+
"ndcg@1": 0.26284055381866905,
|
1498 |
+
"ndcg@5": 0.3879434382781108,
|
1499 |
+
"ndcg@10": 0.42160700212311414,
|
1500 |
+
"precision@1": 0.26284055381866905,
|
1501 |
+
"precision@5": 0.1004019651630192,
|
1502 |
+
"precision@10": 0.060585082626172404,
|
1503 |
+
"recall@1": 0.26284055381866905,
|
1504 |
+
"recall@5": 0.502009825815096,
|
1505 |
+
"recall@10": 0.6058508262617239,
|
1506 |
+
"f1@1": 0.26284055381866905,
|
1507 |
+
"f1@5": 0.16733660860503205,
|
1508 |
+
"f1@10": 0.11015469568394982,
|
1509 |
+
"map@1": 0.26284055381866905,
|
1510 |
+
"map@5": 0.3502047044811672,
|
1511 |
+
"map@10": 0.3641473836817741,
|
1512 |
+
"mrr@1": 0.26284055381866905,
|
1513 |
+
"mrr@5": 0.3502047044811672,
|
1514 |
+
"mrr@10": 0.3641473836817741,
|
1515 |
+
"num_pred": 4478,
|
1516 |
+
"num_data": 4478
|
1517 |
+
},
|
1518 |
+
"YouCook2": {
|
1519 |
+
"hit@1": 0.10632274300094369,
|
1520 |
+
"hit@5": 0.23057565272098143,
|
1521 |
+
"hit@10": 0.2994652406417112,
|
1522 |
+
"ndcg@1": 0.10632274300094369,
|
1523 |
+
"ndcg@5": 0.17009160452701977,
|
1524 |
+
"ndcg@10": 0.19225946072948402,
|
1525 |
+
"precision@1": 0.10632274300094369,
|
1526 |
+
"precision@5": 0.04611513054419629,
|
1527 |
+
"precision@10": 0.02994652406417112,
|
1528 |
+
"recall@1": 0.10632274300094369,
|
1529 |
+
"recall@5": 0.23057565272098143,
|
1530 |
+
"recall@10": 0.2994652406417112,
|
1531 |
+
"f1@1": 0.10632274300094369,
|
1532 |
+
"f1@5": 0.07685855090699382,
|
1533 |
+
"f1@10": 0.054448225571220234,
|
1534 |
+
"map@1": 0.10632274300094369,
|
1535 |
+
"map@5": 0.15018349585823632,
|
1536 |
+
"map@10": 0.15926728980362198,
|
1537 |
+
"mrr@1": 0.10632274300094369,
|
1538 |
+
"mrr@5": 0.15018349585823632,
|
1539 |
+
"mrr@10": 0.15926728980362198,
|
1540 |
+
"num_pred": 3179,
|
1541 |
+
"num_data": 3179
|
1542 |
+
},
|
1543 |
+
"QVHighlight": {
|
1544 |
+
"hit@1": 0.494921514312096,
|
1545 |
+
"hit@5": 0.8679593721144968,
|
1546 |
+
"hit@10": 1.0,
|
1547 |
+
"ndcg@1": 0.494921514312096,
|
1548 |
+
"ndcg@5": 0.6952508060554057,
|
1549 |
+
"ndcg@10": 0.7379681905604544,
|
1550 |
+
"precision@1": 0.494921514312096,
|
1551 |
+
"precision@5": 0.17359187442289936,
|
1552 |
+
"precision@10": 0.09999999999999998,
|
1553 |
+
"recall@1": 0.494921514312096,
|
1554 |
+
"recall@5": 0.8679593721144968,
|
1555 |
+
"recall@10": 1.0,
|
1556 |
+
"f1@1": 0.494921514312096,
|
1557 |
+
"f1@5": 0.2893197907048323,
|
1558 |
+
"f1@10": 0.18181818181818185,
|
1559 |
+
"map@1": 0.494921514312096,
|
1560 |
+
"map@5": 0.6375500153893506,
|
1561 |
+
"map@10": 0.65519830277448,
|
1562 |
+
"mrr@1": 0.494921514312096,
|
1563 |
+
"mrr@5": 0.6375500153893506,
|
1564 |
+
"mrr@10": 0.65519830277448,
|
1565 |
+
"num_pred": 1083,
|
1566 |
+
"num_data": 1083
|
1567 |
+
},
|
1568 |
+
"Charades-STA": {
|
1569 |
+
"hit@1": 0.20220082530949107,
|
1570 |
+
"hit@5": 0.5997248968363136,
|
1571 |
+
"hit@10": 1.0,
|
1572 |
+
"ndcg@1": 0.20220082530949107,
|
1573 |
+
"ndcg@5": 0.3983984877146943,
|
1574 |
+
"ndcg@10": 0.5244775194010131,
|
1575 |
+
"precision@1": 0.20220082530949107,
|
1576 |
+
"precision@5": 0.11994497936726271,
|
1577 |
+
"precision@10": 0.09999999999999998,
|
1578 |
+
"recall@1": 0.20220082530949107,
|
1579 |
+
"recall@5": 0.5997248968363136,
|
1580 |
+
"recall@10": 1.0,
|
1581 |
+
"f1@1": 0.20220082530949107,
|
1582 |
+
"f1@5": 0.19990829894543793,
|
1583 |
+
"f1@10": 0.18181818181818185,
|
1584 |
+
"map@1": 0.20220082530949107,
|
1585 |
+
"map@5": 0.33289775332416327,
|
1586 |
+
"map@10": 0.3829910918975568,
|
1587 |
+
"mrr@1": 0.20220082530949107,
|
1588 |
+
"mrr@5": 0.33289775332416327,
|
1589 |
+
"mrr@10": 0.3829910918975568,
|
1590 |
+
"num_pred": 727,
|
1591 |
+
"num_data": 727
|
1592 |
+
},
|
1593 |
+
"MomentSeeker": {
|
1594 |
+
"hit@1": 0.457,
|
1595 |
+
"hit@5": 0.884,
|
1596 |
+
"hit@10": 1.0,
|
1597 |
+
"ndcg@1": 0.457,
|
1598 |
+
"ndcg@5": 0.6805321190324356,
|
1599 |
+
"ndcg@10": 0.7204304053257706,
|
1600 |
+
"precision@1": 0.457,
|
1601 |
+
"precision@5": 0.1854,
|
1602 |
+
"precision@10": 0.10640000000000004,
|
1603 |
+
"recall@1": 0.4457083333333334,
|
1604 |
+
"recall@5": 0.8787083333333332,
|
1605 |
+
"recall@10": 0.999625,
|
1606 |
+
"f1@1": 0.44886031746031746,
|
1607 |
+
"f1@5": 0.30299178599178606,
|
1608 |
+
"f1@10": 0.19034848484848482,
|
1609 |
+
"map@1": 0.457,
|
1610 |
+
"map@5": 0.612515,
|
1611 |
+
"map@10": 0.6300287037037038,
|
1612 |
+
"mrr@1": 0.457,
|
1613 |
+
"mrr@5": 0.6163,
|
1614 |
+
"mrr@10": 0.6321801587301589,
|
1615 |
+
"num_pred": 1000,
|
1616 |
+
"num_data": 1000
|
1617 |
+
},
|
1618 |
+
"K700": {
|
1619 |
+
"hit@1": 0.234,
|
1620 |
+
"hit@5": 0.486,
|
1621 |
+
"hit@10": 0.572,
|
1622 |
+
"ndcg@1": 0.234,
|
1623 |
+
"ndcg@5": 0.3699192236684553,
|
1624 |
+
"ndcg@10": 0.39783303677678405,
|
1625 |
+
"precision@1": 0.234,
|
1626 |
+
"precision@5": 0.0972,
|
1627 |
+
"precision@10": 0.057199999999999994,
|
1628 |
+
"recall@1": 0.234,
|
1629 |
+
"recall@5": 0.486,
|
1630 |
+
"recall@10": 0.572,
|
1631 |
+
"f1@1": 0.234,
|
1632 |
+
"f1@5": 0.16200000000000003,
|
1633 |
+
"f1@10": 0.10400000000000002,
|
1634 |
+
"map@1": 0.234,
|
1635 |
+
"map@5": 0.33113333333333334,
|
1636 |
+
"map@10": 0.3427166666666666,
|
1637 |
+
"mrr@1": 0.234,
|
1638 |
+
"mrr@5": 0.33113333333333334,
|
1639 |
+
"mrr@10": 0.3427166666666666,
|
1640 |
+
"num_pred": 1000,
|
1641 |
+
"num_data": 1000
|
1642 |
+
},
|
1643 |
+
"SmthSmthV2": {
|
1644 |
+
"hit@1": 0.426,
|
1645 |
+
"hit@5": 0.691,
|
1646 |
+
"hit@10": 0.787,
|
1647 |
+
"ndcg@1": 0.426,
|
1648 |
+
"ndcg@5": 0.5662715930016919,
|
1649 |
+
"ndcg@10": 0.5976793588521978,
|
1650 |
+
"precision@1": 0.426,
|
1651 |
+
"precision@5": 0.1382,
|
1652 |
+
"precision@10": 0.0787,
|
1653 |
+
"recall@1": 0.426,
|
1654 |
+
"recall@5": 0.691,
|
1655 |
+
"recall@10": 0.787,
|
1656 |
+
"f1@1": 0.426,
|
1657 |
+
"f1@5": 0.2303333333333334,
|
1658 |
+
"f1@10": 0.14309090909090913,
|
1659 |
+
"map@1": 0.426,
|
1660 |
+
"map@5": 0.5248666666666667,
|
1661 |
+
"map@10": 0.5380349206349205,
|
1662 |
+
"mrr@1": 0.426,
|
1663 |
+
"mrr@5": 0.5248666666666667,
|
1664 |
+
"mrr@10": 0.5380349206349205,
|
1665 |
+
"num_pred": 1000,
|
1666 |
+
"num_data": 1000
|
1667 |
+
},
|
1668 |
+
"HMDB51": {
|
1669 |
+
"hit@1": 0.407,
|
1670 |
+
"hit@5": 0.64,
|
1671 |
+
"hit@10": 0.712,
|
1672 |
+
"ndcg@1": 0.407,
|
1673 |
+
"ndcg@5": 0.5333263380673475,
|
1674 |
+
"ndcg@10": 0.5566082309108632,
|
1675 |
+
"precision@1": 0.407,
|
1676 |
+
"precision@5": 0.12799999999999997,
|
1677 |
+
"precision@10": 0.07119999999999999,
|
1678 |
+
"recall@1": 0.407,
|
1679 |
+
"recall@5": 0.64,
|
1680 |
+
"recall@10": 0.712,
|
1681 |
+
"f1@1": 0.407,
|
1682 |
+
"f1@5": 0.2133333333333334,
|
1683 |
+
"f1@10": 0.12945454545454546,
|
1684 |
+
"map@1": 0.407,
|
1685 |
+
"map@5": 0.49765,
|
1686 |
+
"map@10": 0.5072710317460318,
|
1687 |
+
"mrr@1": 0.407,
|
1688 |
+
"mrr@5": 0.49765,
|
1689 |
+
"mrr@10": 0.5072710317460318,
|
1690 |
+
"num_pred": 1000,
|
1691 |
+
"num_data": 1000
|
1692 |
+
},
|
1693 |
+
"UCF101": {
|
1694 |
+
"hit@1": 0.6,
|
1695 |
+
"hit@5": 0.833,
|
1696 |
+
"hit@10": 0.887,
|
1697 |
+
"ndcg@1": 0.6,
|
1698 |
+
"ndcg@5": 0.7298423935506184,
|
1699 |
+
"ndcg@10": 0.7473654648933082,
|
1700 |
+
"precision@1": 0.6,
|
1701 |
+
"precision@5": 0.1666,
|
1702 |
+
"precision@10": 0.08870000000000001,
|
1703 |
+
"recall@1": 0.6,
|
1704 |
+
"recall@5": 0.833,
|
1705 |
+
"recall@10": 0.887,
|
1706 |
+
"f1@1": 0.6,
|
1707 |
+
"f1@5": 0.2776666666666667,
|
1708 |
+
"f1@10": 0.1612727272727273,
|
1709 |
+
"map@1": 0.6,
|
1710 |
+
"map@5": 0.69505,
|
1711 |
+
"map@10": 0.7023154761904762,
|
1712 |
+
"mrr@1": 0.6,
|
1713 |
+
"mrr@5": 0.69505,
|
1714 |
+
"mrr@10": 0.7023154761904762,
|
1715 |
+
"num_pred": 1000,
|
1716 |
+
"num_data": 1000
|
1717 |
+
},
|
1718 |
+
"Breakfast": {
|
1719 |
+
"hit@1": 0.15011547344110854,
|
1720 |
+
"hit@5": 0.6397228637413395,
|
1721 |
+
"hit@10": 1.0,
|
1722 |
+
"ndcg@1": 0.15011547344110854,
|
1723 |
+
"ndcg@5": 0.3888398862116513,
|
1724 |
+
"ndcg@10": 0.5075019348255293,
|
1725 |
+
"precision@1": 0.15011547344110854,
|
1726 |
+
"precision@5": 0.12794457274826787,
|
1727 |
+
"precision@10": 0.1,
|
1728 |
+
"recall@1": 0.15011547344110854,
|
1729 |
+
"recall@5": 0.6397228637413395,
|
1730 |
+
"recall@10": 1.0,
|
1731 |
+
"f1@1": 0.15011547344110854,
|
1732 |
+
"f1@5": 0.21324095458044656,
|
1733 |
+
"f1@10": 0.1818181818181818,
|
1734 |
+
"map@1": 0.15011547344110854,
|
1735 |
+
"map@5": 0.30739030023094693,
|
1736 |
+
"map@10": 0.3576175812896368,
|
1737 |
+
"mrr@1": 0.15011547344110854,
|
1738 |
+
"mrr@5": 0.30739030023094693,
|
1739 |
+
"mrr@10": 0.3576175812896368,
|
1740 |
+
"num_pred": 433,
|
1741 |
+
"num_data": 433
|
1742 |
+
},
|
1743 |
+
"MVBench": {
|
1744 |
+
"hit@1": 0.33675,
|
1745 |
+
"hit@5": 1.0,
|
1746 |
+
"hit@10": 1.0,
|
1747 |
+
"ndcg@1": 0.33675,
|
1748 |
+
"ndcg@5": 0.7009772078356213,
|
1749 |
+
"ndcg@10": 0.7009772078356213,
|
1750 |
+
"precision@1": 0.33675,
|
1751 |
+
"precision@5": 0.20000000000000004,
|
1752 |
+
"precision@10": 0.10000000000000002,
|
1753 |
+
"recall@1": 0.33675,
|
1754 |
+
"recall@5": 1.0,
|
1755 |
+
"recall@10": 1.0,
|
1756 |
+
"f1@1": 0.33675,
|
1757 |
+
"f1@5": 0.3333333333333333,
|
1758 |
+
"f1@10": 0.18181818181818182,
|
1759 |
+
"map@1": 0.33675,
|
1760 |
+
"map@5": 0.5999791666666666,
|
1761 |
+
"map@10": 0.5999791666666666,
|
1762 |
+
"mrr@1": 0.33675,
|
1763 |
+
"mrr@5": 0.5999791666666666,
|
1764 |
+
"mrr@10": 0.5999791666666666,
|
1765 |
+
"num_pred": 4000,
|
1766 |
+
"num_data": 4000
|
1767 |
+
},
|
1768 |
+
"Video-MME": {
|
1769 |
+
"hit@1": 0.3048148148148148,
|
1770 |
+
"hit@5": 1.0,
|
1771 |
+
"hit@10": 1.0,
|
1772 |
+
"ndcg@1": 0.3048148148148148,
|
1773 |
+
"ndcg@5": 0.6736313794424496,
|
1774 |
+
"ndcg@10": 0.6736313794424496,
|
1775 |
+
"precision@1": 0.3048148148148148,
|
1776 |
+
"precision@5": 0.19999999999999996,
|
1777 |
+
"precision@10": 0.09999999999999998,
|
1778 |
+
"recall@1": 0.3048148148148148,
|
1779 |
+
"recall@5": 1.0,
|
1780 |
+
"recall@10": 1.0,
|
1781 |
+
"f1@1": 0.3048148148148148,
|
1782 |
+
"f1@5": 0.3333333333333334,
|
1783 |
+
"f1@10": 0.18181818181818188,
|
1784 |
+
"map@1": 0.3048148148148148,
|
1785 |
+
"map@5": 0.5645061728395061,
|
1786 |
+
"map@10": 0.5645061728395061,
|
1787 |
+
"mrr@1": 0.3048148148148148,
|
1788 |
+
"mrr@5": 0.5645061728395061,
|
1789 |
+
"mrr@10": 0.5645061728395061,
|
1790 |
+
"num_pred": 2700,
|
1791 |
+
"num_data": 2700
|
1792 |
+
},
|
1793 |
+
"NExTQA": {
|
1794 |
+
"hit@1": 0.20924801494628678,
|
1795 |
+
"hit@5": 1.0,
|
1796 |
+
"hit@10": 1.0,
|
1797 |
+
"ndcg@1": 0.20924801494628678,
|
1798 |
+
"ndcg@5": 0.6098628990350217,
|
1799 |
+
"ndcg@10": 0.6098628990350217,
|
1800 |
+
"precision@1": 0.20924801494628678,
|
1801 |
+
"precision@5": 0.20000000000000004,
|
1802 |
+
"precision@10": 0.10000000000000002,
|
1803 |
+
"recall@1": 0.20924801494628678,
|
1804 |
+
"recall@5": 1.0,
|
1805 |
+
"recall@10": 1.0,
|
1806 |
+
"f1@1": 0.20924801494628678,
|
1807 |
+
"f1@5": 0.3333333333333333,
|
1808 |
+
"f1@10": 0.1818181818181818,
|
1809 |
+
"map@1": 0.20924801494628678,
|
1810 |
+
"map@5": 0.4819321189475323,
|
1811 |
+
"map@10": 0.4819321189475323,
|
1812 |
+
"mrr@1": 0.20924801494628678,
|
1813 |
+
"mrr@5": 0.48175696714930716,
|
1814 |
+
"mrr@10": 0.48175696714930716,
|
1815 |
+
"num_pred": 8564,
|
1816 |
+
"num_data": 8564
|
1817 |
+
},
|
1818 |
+
"EgoSchema": {
|
1819 |
+
"hit@1": 0.344,
|
1820 |
+
"hit@5": 1.0,
|
1821 |
+
"hit@10": 1.0,
|
1822 |
+
"ndcg@1": 0.344,
|
1823 |
+
"ndcg@5": 0.6824282286336308,
|
1824 |
+
"ndcg@10": 0.6824282286336308,
|
1825 |
+
"precision@1": 0.344,
|
1826 |
+
"precision@5": 0.2,
|
1827 |
+
"precision@10": 0.1,
|
1828 |
+
"recall@1": 0.344,
|
1829 |
+
"recall@5": 1.0,
|
1830 |
+
"recall@10": 1.0,
|
1831 |
+
"f1@1": 0.344,
|
1832 |
+
"f1@5": 0.3333333333333333,
|
1833 |
+
"f1@10": 0.18181818181818182,
|
1834 |
+
"map@1": 0.344,
|
1835 |
+
"map@5": 0.5776333333333333,
|
1836 |
+
"map@10": 0.5776333333333333,
|
1837 |
+
"mrr@1": 0.344,
|
1838 |
+
"mrr@5": 0.5776333333333333,
|
1839 |
+
"mrr@10": 0.5776333333333333,
|
1840 |
+
"num_pred": 500,
|
1841 |
+
"num_data": 500
|
1842 |
+
},
|
1843 |
+
"ActivityNetQA": {
|
1844 |
+
"hit@1": 0.523,
|
1845 |
+
"hit@5": 1.0,
|
1846 |
+
"hit@10": 1.0,
|
1847 |
+
"ndcg@1": 0.523,
|
1848 |
+
"ndcg@5": 0.8239534924535853,
|
1849 |
+
"ndcg@10": 0.8239534924535853,
|
1850 |
+
"precision@1": 0.523,
|
1851 |
+
"precision@5": 0.20000000000000004,
|
1852 |
+
"precision@10": 0.10000000000000002,
|
1853 |
+
"recall@1": 0.523,
|
1854 |
+
"recall@5": 1.0,
|
1855 |
+
"recall@10": 1.0,
|
1856 |
+
"f1@1": 0.523,
|
1857 |
+
"f1@5": 0.3333333333333333,
|
1858 |
+
"f1@10": 0.18181818181818182,
|
1859 |
+
"map@1": 0.523,
|
1860 |
+
"map@5": 0.7615,
|
1861 |
+
"map@10": 0.7615,
|
1862 |
+
"mrr@1": 0.523,
|
1863 |
+
"mrr@5": 0.7615,
|
1864 |
+
"mrr@10": 0.7615,
|
1865 |
+
"num_pred": 1000,
|
1866 |
+
"num_data": 1000
|
1867 |
+
}
|
1868 |
+
}
|
1869 |
+
}
|
1870 |
+
}
|
Scores/VLM2Vec-Qwen2-VL-V2.1-scores_report.json
ADDED
@@ -0,0 +1,1702 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"metadata": {
|
3 |
+
"model_name": "VLM2Vec-Qwen2-VL-V2.1",
|
4 |
+
"model_size": 2,
|
5 |
+
"embedding_dimension": null,
|
6 |
+
"max_length_tokens": null,
|
7 |
+
"model_release_date": "2025-05-15",
|
8 |
+
"data_source": "",
|
9 |
+
"url": "",
|
10 |
+
"report_generated_date": "2025-05-30T06:33:26.437452"
|
11 |
+
},
|
12 |
+
"metrics": {
|
13 |
+
"image": {
|
14 |
+
"ImageNet-1K": {
|
15 |
+
"hit@1": 0.786,
|
16 |
+
"hit@5": 0.951,
|
17 |
+
"hit@10": 0.974,
|
18 |
+
"ndcg@1": 0.786,
|
19 |
+
"ndcg@5": 0.8793557596567982,
|
20 |
+
"ndcg@10": 0.8869605540910463,
|
21 |
+
"precision@1": 0.786,
|
22 |
+
"precision@5": 0.1902,
|
23 |
+
"precision@10": 0.0974,
|
24 |
+
"recall@1": 0.786,
|
25 |
+
"recall@5": 0.951,
|
26 |
+
"recall@10": 0.974,
|
27 |
+
"f1@1": 0.786,
|
28 |
+
"f1@5": 0.317,
|
29 |
+
"f1@10": 0.1770909090909091,
|
30 |
+
"map@1": 0.786,
|
31 |
+
"map@5": 0.855,
|
32 |
+
"map@10": 0.8582321428571429,
|
33 |
+
"mrr@1": 0.786,
|
34 |
+
"mrr@5": 0.855,
|
35 |
+
"mrr@10": 0.8582321428571429,
|
36 |
+
"num_pred": 1000,
|
37 |
+
"num_data": 1000
|
38 |
+
},
|
39 |
+
"N24News": {
|
40 |
+
"hit@1": 0.765,
|
41 |
+
"hit@5": 0.964,
|
42 |
+
"hit@10": 0.989,
|
43 |
+
"ndcg@1": 0.765,
|
44 |
+
"ndcg@5": 0.8777475188437514,
|
45 |
+
"ndcg@10": 0.8857877852100778,
|
46 |
+
"precision@1": 0.765,
|
47 |
+
"precision@5": 0.1928,
|
48 |
+
"precision@10": 0.0989,
|
49 |
+
"recall@1": 0.765,
|
50 |
+
"recall@5": 0.964,
|
51 |
+
"recall@10": 0.989,
|
52 |
+
"f1@1": 0.765,
|
53 |
+
"f1@5": 0.32133333333333336,
|
54 |
+
"f1@10": 0.1798181818181818,
|
55 |
+
"map@1": 0.765,
|
56 |
+
"map@5": 0.8484166666666666,
|
57 |
+
"map@10": 0.8517059523809524,
|
58 |
+
"mrr@1": 0.765,
|
59 |
+
"mrr@5": 0.8484166666666666,
|
60 |
+
"mrr@10": 0.8517059523809524,
|
61 |
+
"num_pred": 1000,
|
62 |
+
"num_data": 1000
|
63 |
+
},
|
64 |
+
"HatefulMemes": {
|
65 |
+
"hit@1": 0.539,
|
66 |
+
"hit@5": 1.0,
|
67 |
+
"hit@10": 1.0,
|
68 |
+
"ndcg@1": 0.539,
|
69 |
+
"ndcg@5": 0.829858616396442,
|
70 |
+
"ndcg@10": 0.829858616396442,
|
71 |
+
"precision@1": 0.539,
|
72 |
+
"precision@5": 0.20000000000000004,
|
73 |
+
"precision@10": 0.10000000000000002,
|
74 |
+
"recall@1": 0.539,
|
75 |
+
"recall@5": 1.0,
|
76 |
+
"recall@10": 1.0,
|
77 |
+
"f1@1": 0.539,
|
78 |
+
"f1@5": 0.3333333333333333,
|
79 |
+
"f1@10": 0.18181818181818182,
|
80 |
+
"map@1": 0.539,
|
81 |
+
"map@5": 0.7695,
|
82 |
+
"map@10": 0.7695,
|
83 |
+
"mrr@1": 0.539,
|
84 |
+
"mrr@5": 0.7695,
|
85 |
+
"mrr@10": 0.7695,
|
86 |
+
"num_pred": 1000,
|
87 |
+
"num_data": 1000
|
88 |
+
},
|
89 |
+
"VOC2007": {
|
90 |
+
"hit@1": 0.856,
|
91 |
+
"hit@5": 0.97,
|
92 |
+
"hit@10": 0.988,
|
93 |
+
"ndcg@1": 0.856,
|
94 |
+
"ndcg@5": 0.9206875351116595,
|
95 |
+
"ndcg@10": 0.9266647673940863,
|
96 |
+
"precision@1": 0.856,
|
97 |
+
"precision@5": 0.19400000000000003,
|
98 |
+
"precision@10": 0.09880000000000001,
|
99 |
+
"recall@1": 0.856,
|
100 |
+
"recall@5": 0.97,
|
101 |
+
"recall@10": 0.988,
|
102 |
+
"f1@1": 0.856,
|
103 |
+
"f1@5": 0.32333333333333336,
|
104 |
+
"f1@10": 0.17963636363636362,
|
105 |
+
"map@1": 0.856,
|
106 |
+
"map@5": 0.9038999999999999,
|
107 |
+
"map@10": 0.906457142857143,
|
108 |
+
"mrr@1": 0.856,
|
109 |
+
"mrr@5": 0.9038999999999999,
|
110 |
+
"mrr@10": 0.906457142857143,
|
111 |
+
"num_pred": 1000,
|
112 |
+
"num_data": 1000
|
113 |
+
},
|
114 |
+
"SUN397": {
|
115 |
+
"hit@1": 0.652,
|
116 |
+
"hit@5": 0.898,
|
117 |
+
"hit@10": 0.947,
|
118 |
+
"ndcg@1": 0.652,
|
119 |
+
"ndcg@5": 0.7878784740466801,
|
120 |
+
"ndcg@10": 0.8042132482472114,
|
121 |
+
"precision@1": 0.652,
|
122 |
+
"precision@5": 0.1796,
|
123 |
+
"precision@10": 0.09470000000000002,
|
124 |
+
"recall@1": 0.652,
|
125 |
+
"recall@5": 0.898,
|
126 |
+
"recall@10": 0.947,
|
127 |
+
"f1@1": 0.652,
|
128 |
+
"f1@5": 0.2993333333333334,
|
129 |
+
"f1@10": 0.17218181818181819,
|
130 |
+
"map@1": 0.652,
|
131 |
+
"map@5": 0.75075,
|
132 |
+
"map@10": 0.757777380952381,
|
133 |
+
"mrr@1": 0.652,
|
134 |
+
"mrr@5": 0.75075,
|
135 |
+
"mrr@10": 0.757777380952381,
|
136 |
+
"num_pred": 1000,
|
137 |
+
"num_data": 1000
|
138 |
+
},
|
139 |
+
"Place365": {
|
140 |
+
"hit@1": 0.338,
|
141 |
+
"hit@5": 0.603,
|
142 |
+
"hit@10": 0.699,
|
143 |
+
"ndcg@1": 0.338,
|
144 |
+
"ndcg@5": 0.4803878136483831,
|
145 |
+
"ndcg@10": 0.5115497912882357,
|
146 |
+
"precision@1": 0.338,
|
147 |
+
"precision@5": 0.12059999999999998,
|
148 |
+
"precision@10": 0.0699,
|
149 |
+
"recall@1": 0.338,
|
150 |
+
"recall@5": 0.603,
|
151 |
+
"recall@10": 0.699,
|
152 |
+
"f1@1": 0.338,
|
153 |
+
"f1@5": 0.20100000000000007,
|
154 |
+
"f1@10": 0.12709090909090912,
|
155 |
+
"map@1": 0.338,
|
156 |
+
"map@5": 0.43948333333333334,
|
157 |
+
"map@10": 0.452411507936508,
|
158 |
+
"mrr@1": 0.338,
|
159 |
+
"mrr@5": 0.43948333333333334,
|
160 |
+
"mrr@10": 0.452411507936508,
|
161 |
+
"num_pred": 1000,
|
162 |
+
"num_data": 1000
|
163 |
+
},
|
164 |
+
"ImageNet-A": {
|
165 |
+
"hit@1": 0.47,
|
166 |
+
"hit@5": 0.709,
|
167 |
+
"hit@10": 0.787,
|
168 |
+
"ndcg@1": 0.47,
|
169 |
+
"ndcg@5": 0.5994393029556508,
|
170 |
+
"ndcg@10": 0.6248293448191466,
|
171 |
+
"precision@1": 0.47,
|
172 |
+
"precision@5": 0.1418,
|
173 |
+
"precision@10": 0.0787,
|
174 |
+
"recall@1": 0.47,
|
175 |
+
"recall@5": 0.709,
|
176 |
+
"recall@10": 0.787,
|
177 |
+
"f1@1": 0.47,
|
178 |
+
"f1@5": 0.2363333333333334,
|
179 |
+
"f1@10": 0.14309090909090913,
|
180 |
+
"map@1": 0.47,
|
181 |
+
"map@5": 0.5626833333333334,
|
182 |
+
"map@10": 0.5732492063492064,
|
183 |
+
"mrr@1": 0.47,
|
184 |
+
"mrr@5": 0.5626833333333334,
|
185 |
+
"mrr@10": 0.5732492063492064,
|
186 |
+
"num_pred": 1000,
|
187 |
+
"num_data": 1000
|
188 |
+
},
|
189 |
+
"ImageNet-R": {
|
190 |
+
"hit@1": 0.814,
|
191 |
+
"hit@5": 0.955,
|
192 |
+
"hit@10": 0.972,
|
193 |
+
"ndcg@1": 0.814,
|
194 |
+
"ndcg@5": 0.8898798613976422,
|
195 |
+
"ndcg@10": 0.8954397133930482,
|
196 |
+
"precision@1": 0.814,
|
197 |
+
"precision@5": 0.19100000000000006,
|
198 |
+
"precision@10": 0.09720000000000002,
|
199 |
+
"recall@1": 0.814,
|
200 |
+
"recall@5": 0.955,
|
201 |
+
"recall@10": 0.972,
|
202 |
+
"f1@1": 0.814,
|
203 |
+
"f1@5": 0.31833333333333336,
|
204 |
+
"f1@10": 0.17672727272727268,
|
205 |
+
"map@1": 0.814,
|
206 |
+
"map@5": 0.868,
|
207 |
+
"map@10": 0.8703309523809524,
|
208 |
+
"mrr@1": 0.814,
|
209 |
+
"mrr@5": 0.868,
|
210 |
+
"mrr@10": 0.8703309523809524,
|
211 |
+
"num_pred": 1000,
|
212 |
+
"num_data": 1000
|
213 |
+
},
|
214 |
+
"ObjectNet": {
|
215 |
+
"hit@1": 0.637,
|
216 |
+
"hit@5": 0.832,
|
217 |
+
"hit@10": 0.888,
|
218 |
+
"ndcg@1": 0.637,
|
219 |
+
"ndcg@5": 0.7442422525944117,
|
220 |
+
"ndcg@10": 0.7624181370332057,
|
221 |
+
"precision@1": 0.637,
|
222 |
+
"precision@5": 0.1664,
|
223 |
+
"precision@10": 0.08880000000000002,
|
224 |
+
"recall@1": 0.637,
|
225 |
+
"recall@5": 0.832,
|
226 |
+
"recall@10": 0.888,
|
227 |
+
"f1@1": 0.637,
|
228 |
+
"f1@5": 0.2773333333333334,
|
229 |
+
"f1@10": 0.16145454545454546,
|
230 |
+
"map@1": 0.637,
|
231 |
+
"map@5": 0.7148,
|
232 |
+
"map@10": 0.7223412698412699,
|
233 |
+
"mrr@1": 0.637,
|
234 |
+
"mrr@5": 0.7148,
|
235 |
+
"mrr@10": 0.7223412698412699,
|
236 |
+
"num_pred": 1000,
|
237 |
+
"num_data": 1000
|
238 |
+
},
|
239 |
+
"Country211": {
|
240 |
+
"hit@1": 0.236,
|
241 |
+
"hit@5": 0.435,
|
242 |
+
"hit@10": 0.522,
|
243 |
+
"ndcg@1": 0.236,
|
244 |
+
"ndcg@5": 0.3391707021175824,
|
245 |
+
"ndcg@10": 0.3671683360049007,
|
246 |
+
"precision@1": 0.236,
|
247 |
+
"precision@5": 0.087,
|
248 |
+
"precision@10": 0.0522,
|
249 |
+
"recall@1": 0.236,
|
250 |
+
"recall@5": 0.435,
|
251 |
+
"recall@10": 0.522,
|
252 |
+
"f1@1": 0.236,
|
253 |
+
"f1@5": 0.14500000000000005,
|
254 |
+
"f1@10": 0.09490909090909093,
|
255 |
+
"map@1": 0.236,
|
256 |
+
"map@5": 0.30751666666666666,
|
257 |
+
"map@10": 0.31899325396825395,
|
258 |
+
"mrr@1": 0.236,
|
259 |
+
"mrr@5": 0.30751666666666666,
|
260 |
+
"mrr@10": 0.31899325396825395,
|
261 |
+
"num_pred": 1000,
|
262 |
+
"num_data": 1000
|
263 |
+
},
|
264 |
+
"OK-VQA": {
|
265 |
+
"hit@1": 0.53,
|
266 |
+
"hit@5": 0.788,
|
267 |
+
"hit@10": 0.858,
|
268 |
+
"ndcg@1": 0.53,
|
269 |
+
"ndcg@5": 0.6698161425813521,
|
270 |
+
"ndcg@10": 0.6927084829279392,
|
271 |
+
"precision@1": 0.53,
|
272 |
+
"precision@5": 0.15760000000000002,
|
273 |
+
"precision@10": 0.08580000000000002,
|
274 |
+
"recall@1": 0.53,
|
275 |
+
"recall@5": 0.788,
|
276 |
+
"recall@10": 0.858,
|
277 |
+
"f1@1": 0.53,
|
278 |
+
"f1@5": 0.2626666666666667,
|
279 |
+
"f1@10": 0.156,
|
280 |
+
"map@1": 0.53,
|
281 |
+
"map@5": 0.6302166666666666,
|
282 |
+
"map@10": 0.6398103174603175,
|
283 |
+
"mrr@1": 0.53,
|
284 |
+
"mrr@5": 0.6302166666666666,
|
285 |
+
"mrr@10": 0.6398103174603175,
|
286 |
+
"num_pred": 1000,
|
287 |
+
"num_data": 1000
|
288 |
+
},
|
289 |
+
"A-OKVQA": {
|
290 |
+
"hit@1": 0.438,
|
291 |
+
"hit@5": 0.752,
|
292 |
+
"hit@10": 0.832,
|
293 |
+
"ndcg@1": 0.438,
|
294 |
+
"ndcg@5": 0.6065021222125251,
|
295 |
+
"ndcg@10": 0.6325055288198025,
|
296 |
+
"precision@1": 0.438,
|
297 |
+
"precision@5": 0.1504,
|
298 |
+
"precision@10": 0.0832,
|
299 |
+
"recall@1": 0.438,
|
300 |
+
"recall@5": 0.752,
|
301 |
+
"recall@10": 0.832,
|
302 |
+
"f1@1": 0.438,
|
303 |
+
"f1@5": 0.2506666666666667,
|
304 |
+
"f1@10": 0.15127272727272728,
|
305 |
+
"map@1": 0.438,
|
306 |
+
"map@5": 0.5579666666666667,
|
307 |
+
"map@10": 0.5687769841269841,
|
308 |
+
"mrr@1": 0.438,
|
309 |
+
"mrr@5": 0.5579666666666667,
|
310 |
+
"mrr@10": 0.5687769841269841,
|
311 |
+
"num_pred": 1000,
|
312 |
+
"num_data": 1000
|
313 |
+
},
|
314 |
+
"DocVQA": {
|
315 |
+
"hit@1": 0.896,
|
316 |
+
"hit@5": 0.965,
|
317 |
+
"hit@10": 0.976,
|
318 |
+
"ndcg@1": 0.896,
|
319 |
+
"ndcg@5": 0.9360552871316068,
|
320 |
+
"ndcg@10": 0.9395738155061311,
|
321 |
+
"precision@1": 0.896,
|
322 |
+
"precision@5": 0.19300000000000003,
|
323 |
+
"precision@10": 0.09760000000000002,
|
324 |
+
"recall@1": 0.896,
|
325 |
+
"recall@5": 0.965,
|
326 |
+
"recall@10": 0.976,
|
327 |
+
"f1@1": 0.896,
|
328 |
+
"f1@5": 0.3216666666666667,
|
329 |
+
"f1@10": 0.17745454545454545,
|
330 |
+
"map@1": 0.896,
|
331 |
+
"map@5": 0.9261166666666667,
|
332 |
+
"map@10": 0.9275448412698413,
|
333 |
+
"mrr@1": 0.896,
|
334 |
+
"mrr@5": 0.9261166666666667,
|
335 |
+
"mrr@10": 0.9275448412698413,
|
336 |
+
"num_pred": 1000,
|
337 |
+
"num_data": 1000
|
338 |
+
},
|
339 |
+
"InfographicsVQA": {
|
340 |
+
"hit@1": 0.559,
|
341 |
+
"hit@5": 0.793,
|
342 |
+
"hit@10": 0.864,
|
343 |
+
"ndcg@1": 0.559,
|
344 |
+
"ndcg@5": 0.685957267820919,
|
345 |
+
"ndcg@10": 0.708607344773437,
|
346 |
+
"precision@1": 0.559,
|
347 |
+
"precision@5": 0.1586,
|
348 |
+
"precision@10": 0.0864,
|
349 |
+
"recall@1": 0.559,
|
350 |
+
"recall@5": 0.793,
|
351 |
+
"recall@10": 0.864,
|
352 |
+
"f1@1": 0.559,
|
353 |
+
"f1@5": 0.26433333333333336,
|
354 |
+
"f1@10": 0.1570909090909091,
|
355 |
+
"map@1": 0.559,
|
356 |
+
"map@5": 0.6501499999999999,
|
357 |
+
"map@10": 0.6593198412698412,
|
358 |
+
"mrr@1": 0.559,
|
359 |
+
"mrr@5": 0.6501499999999999,
|
360 |
+
"mrr@10": 0.6593198412698412,
|
361 |
+
"num_pred": 1000,
|
362 |
+
"num_data": 1000
|
363 |
+
},
|
364 |
+
"ChartQA": {
|
365 |
+
"hit@1": 0.461,
|
366 |
+
"hit@5": 0.671,
|
367 |
+
"hit@10": 0.753,
|
368 |
+
"ndcg@1": 0.461,
|
369 |
+
"ndcg@5": 0.5739689048241495,
|
370 |
+
"ndcg@10": 0.6004441942084271,
|
371 |
+
"precision@1": 0.461,
|
372 |
+
"precision@5": 0.1342,
|
373 |
+
"precision@10": 0.0753,
|
374 |
+
"recall@1": 0.461,
|
375 |
+
"recall@5": 0.671,
|
376 |
+
"recall@10": 0.753,
|
377 |
+
"f1@1": 0.461,
|
378 |
+
"f1@5": 0.22366666666666668,
|
379 |
+
"f1@10": 0.13690909090909092,
|
380 |
+
"map@1": 0.461,
|
381 |
+
"map@5": 0.5415833333333333,
|
382 |
+
"map@10": 0.5524876984126984,
|
383 |
+
"mrr@1": 0.461,
|
384 |
+
"mrr@5": 0.5415833333333333,
|
385 |
+
"mrr@10": 0.5524876984126984,
|
386 |
+
"num_pred": 1000,
|
387 |
+
"num_data": 1000
|
388 |
+
},
|
389 |
+
"Visual7W": {
|
390 |
+
"hit@1": 0.474,
|
391 |
+
"hit@5": 0.855,
|
392 |
+
"hit@10": 0.934,
|
393 |
+
"ndcg@1": 0.474,
|
394 |
+
"ndcg@5": 0.6817991379902264,
|
395 |
+
"ndcg@10": 0.7077098354826533,
|
396 |
+
"precision@1": 0.474,
|
397 |
+
"precision@5": 0.171,
|
398 |
+
"precision@10": 0.09340000000000001,
|
399 |
+
"recall@1": 0.474,
|
400 |
+
"recall@5": 0.855,
|
401 |
+
"recall@10": 0.934,
|
402 |
+
"f1@1": 0.474,
|
403 |
+
"f1@5": 0.285,
|
404 |
+
"f1@10": 0.1698181818181818,
|
405 |
+
"map@1": 0.474,
|
406 |
+
"map@5": 0.6236833333333334,
|
407 |
+
"map@10": 0.6345873015873017,
|
408 |
+
"mrr@1": 0.474,
|
409 |
+
"mrr@5": 0.6236833333333334,
|
410 |
+
"mrr@10": 0.6345873015873017,
|
411 |
+
"num_pred": 1000,
|
412 |
+
"num_data": 1000
|
413 |
+
},
|
414 |
+
"ScienceQA": {
|
415 |
+
"hit@1": 0.37,
|
416 |
+
"hit@5": 0.659,
|
417 |
+
"hit@10": 0.79,
|
418 |
+
"ndcg@1": 0.37,
|
419 |
+
"ndcg@5": 0.5269886261590063,
|
420 |
+
"ndcg@10": 0.5692567171878673,
|
421 |
+
"precision@1": 0.37,
|
422 |
+
"precision@5": 0.1318,
|
423 |
+
"precision@10": 0.079,
|
424 |
+
"recall@1": 0.37,
|
425 |
+
"recall@5": 0.659,
|
426 |
+
"recall@10": 0.79,
|
427 |
+
"f1@1": 0.37,
|
428 |
+
"f1@5": 0.21966666666666673,
|
429 |
+
"f1@10": 0.14363636363636365,
|
430 |
+
"map@1": 0.37,
|
431 |
+
"map@5": 0.4827833333333333,
|
432 |
+
"map@10": 0.5001579365079365,
|
433 |
+
"mrr@1": 0.37,
|
434 |
+
"mrr@5": 0.4827833333333333,
|
435 |
+
"mrr@10": 0.5001579365079365,
|
436 |
+
"num_pred": 1000,
|
437 |
+
"num_data": 1000
|
438 |
+
},
|
439 |
+
"VizWiz": {
|
440 |
+
"hit@1": 0.422,
|
441 |
+
"hit@5": 0.617,
|
442 |
+
"hit@10": 0.681,
|
443 |
+
"ndcg@1": 0.422,
|
444 |
+
"ndcg@5": 0.5254860929618814,
|
445 |
+
"ndcg@10": 0.5461605768075588,
|
446 |
+
"precision@1": 0.422,
|
447 |
+
"precision@5": 0.1234,
|
448 |
+
"precision@10": 0.0681,
|
449 |
+
"recall@1": 0.422,
|
450 |
+
"recall@5": 0.617,
|
451 |
+
"recall@10": 0.681,
|
452 |
+
"f1@1": 0.422,
|
453 |
+
"f1@5": 0.20566666666666672,
|
454 |
+
"f1@10": 0.12381818181818184,
|
455 |
+
"map@1": 0.422,
|
456 |
+
"map@5": 0.4950833333333334,
|
457 |
+
"map@10": 0.503597619047619,
|
458 |
+
"mrr@1": 0.422,
|
459 |
+
"mrr@5": 0.4950833333333334,
|
460 |
+
"mrr@10": 0.503597619047619,
|
461 |
+
"num_pred": 1000,
|
462 |
+
"num_data": 1000
|
463 |
+
},
|
464 |
+
"GQA": {
|
465 |
+
"hit@1": 0.595,
|
466 |
+
"hit@5": 0.835,
|
467 |
+
"hit@10": 0.904,
|
468 |
+
"ndcg@1": 0.595,
|
469 |
+
"ndcg@5": 0.7203437271941817,
|
470 |
+
"ndcg@10": 0.7425042508341791,
|
471 |
+
"precision@1": 0.595,
|
472 |
+
"precision@5": 0.167,
|
473 |
+
"precision@10": 0.09040000000000002,
|
474 |
+
"recall@1": 0.595,
|
475 |
+
"recall@5": 0.835,
|
476 |
+
"recall@10": 0.904,
|
477 |
+
"f1@1": 0.595,
|
478 |
+
"f1@5": 0.2783333333333334,
|
479 |
+
"f1@10": 0.16436363636363638,
|
480 |
+
"map@1": 0.595,
|
481 |
+
"map@5": 0.6824,
|
482 |
+
"map@10": 0.691459126984127,
|
483 |
+
"mrr@1": 0.595,
|
484 |
+
"mrr@5": 0.6824,
|
485 |
+
"mrr@10": 0.691459126984127,
|
486 |
+
"num_pred": 1000,
|
487 |
+
"num_data": 1000
|
488 |
+
},
|
489 |
+
"TextVQA": {
|
490 |
+
"hit@1": 0.714,
|
491 |
+
"hit@5": 0.828,
|
492 |
+
"hit@10": 0.861,
|
493 |
+
"ndcg@1": 0.714,
|
494 |
+
"ndcg@5": 0.7759883374392802,
|
495 |
+
"ndcg@10": 0.7866798717275013,
|
496 |
+
"precision@1": 0.714,
|
497 |
+
"precision@5": 0.1656,
|
498 |
+
"precision@10": 0.08610000000000001,
|
499 |
+
"recall@1": 0.714,
|
500 |
+
"recall@5": 0.828,
|
501 |
+
"recall@10": 0.861,
|
502 |
+
"f1@1": 0.714,
|
503 |
+
"f1@5": 0.276,
|
504 |
+
"f1@10": 0.15654545454545457,
|
505 |
+
"map@1": 0.714,
|
506 |
+
"map@5": 0.7585666666666666,
|
507 |
+
"map@10": 0.7629837301587302,
|
508 |
+
"mrr@1": 0.714,
|
509 |
+
"mrr@5": 0.7585666666666666,
|
510 |
+
"mrr@10": 0.7629837301587302,
|
511 |
+
"num_pred": 1000,
|
512 |
+
"num_data": 1000
|
513 |
+
},
|
514 |
+
"VisDial": {
|
515 |
+
"hit@1": 0.786,
|
516 |
+
"hit@5": 0.94,
|
517 |
+
"hit@10": 0.974,
|
518 |
+
"ndcg@1": 0.786,
|
519 |
+
"ndcg@5": 0.8712201058279702,
|
520 |
+
"ndcg@10": 0.8822886741609149,
|
521 |
+
"precision@1": 0.786,
|
522 |
+
"precision@5": 0.18800000000000003,
|
523 |
+
"precision@10": 0.09740000000000001,
|
524 |
+
"recall@1": 0.786,
|
525 |
+
"recall@5": 0.94,
|
526 |
+
"recall@10": 0.974,
|
527 |
+
"f1@1": 0.786,
|
528 |
+
"f1@5": 0.31333333333333335,
|
529 |
+
"f1@10": 0.17709090909090908,
|
530 |
+
"map@1": 0.786,
|
531 |
+
"map@5": 0.84805,
|
532 |
+
"map@10": 0.8526611111111112,
|
533 |
+
"mrr@1": 0.786,
|
534 |
+
"mrr@5": 0.84805,
|
535 |
+
"mrr@10": 0.8526611111111112,
|
536 |
+
"num_pred": 1000,
|
537 |
+
"num_data": 1000
|
538 |
+
},
|
539 |
+
"CIRR": {
|
540 |
+
"hit@1": 0.49,
|
541 |
+
"hit@5": 0.877,
|
542 |
+
"hit@10": 0.932,
|
543 |
+
"ndcg@1": 0.49,
|
544 |
+
"ndcg@5": 0.7047483751869181,
|
545 |
+
"ndcg@10": 0.7228061409716358,
|
546 |
+
"precision@1": 0.49,
|
547 |
+
"precision@5": 0.17540000000000003,
|
548 |
+
"precision@10": 0.09320000000000002,
|
549 |
+
"recall@1": 0.49,
|
550 |
+
"recall@5": 0.877,
|
551 |
+
"recall@10": 0.932,
|
552 |
+
"f1@1": 0.49,
|
553 |
+
"f1@5": 0.2923333333333334,
|
554 |
+
"f1@10": 0.16945454545454544,
|
555 |
+
"map@1": 0.49,
|
556 |
+
"map@5": 0.6466333333333333,
|
557 |
+
"map@10": 0.6542412698412697,
|
558 |
+
"mrr@1": 0.49,
|
559 |
+
"mrr@5": 0.6466333333333333,
|
560 |
+
"mrr@10": 0.6542412698412697,
|
561 |
+
"num_pred": 1000,
|
562 |
+
"num_data": 1000
|
563 |
+
},
|
564 |
+
"VisualNews_t2i": {
|
565 |
+
"hit@1": 0.726,
|
566 |
+
"hit@5": 0.887,
|
567 |
+
"hit@10": 0.918,
|
568 |
+
"ndcg@1": 0.726,
|
569 |
+
"ndcg@5": 0.8129961329430482,
|
570 |
+
"ndcg@10": 0.8230809979107395,
|
571 |
+
"precision@1": 0.726,
|
572 |
+
"precision@5": 0.1774,
|
573 |
+
"precision@10": 0.0918,
|
574 |
+
"recall@1": 0.726,
|
575 |
+
"recall@5": 0.887,
|
576 |
+
"recall@10": 0.918,
|
577 |
+
"f1@1": 0.726,
|
578 |
+
"f1@5": 0.29566666666666674,
|
579 |
+
"f1@10": 0.1669090909090909,
|
580 |
+
"map@1": 0.726,
|
581 |
+
"map@5": 0.7882833333333333,
|
582 |
+
"map@10": 0.7924817460317459,
|
583 |
+
"mrr@1": 0.726,
|
584 |
+
"mrr@5": 0.7882833333333333,
|
585 |
+
"mrr@10": 0.7924817460317459,
|
586 |
+
"num_pred": 1000,
|
587 |
+
"num_data": 1000
|
588 |
+
},
|
589 |
+
"VisualNews_i2t": {
|
590 |
+
"hit@1": 0.723,
|
591 |
+
"hit@5": 0.881,
|
592 |
+
"hit@10": 0.919,
|
593 |
+
"ndcg@1": 0.723,
|
594 |
+
"ndcg@5": 0.8092182983515347,
|
595 |
+
"ndcg@10": 0.8217076538993346,
|
596 |
+
"precision@1": 0.723,
|
597 |
+
"precision@5": 0.1762,
|
598 |
+
"precision@10": 0.09190000000000001,
|
599 |
+
"recall@1": 0.723,
|
600 |
+
"recall@5": 0.881,
|
601 |
+
"recall@10": 0.919,
|
602 |
+
"f1@1": 0.723,
|
603 |
+
"f1@5": 0.2936666666666667,
|
604 |
+
"f1@10": 0.1670909090909091,
|
605 |
+
"map@1": 0.723,
|
606 |
+
"map@5": 0.78515,
|
607 |
+
"map@10": 0.7904230158730159,
|
608 |
+
"mrr@1": 0.723,
|
609 |
+
"mrr@5": 0.78515,
|
610 |
+
"mrr@10": 0.7904230158730159,
|
611 |
+
"num_pred": 1000,
|
612 |
+
"num_data": 1000
|
613 |
+
},
|
614 |
+
"MSCOCO_t2i": {
|
615 |
+
"hit@1": 0.714,
|
616 |
+
"hit@5": 0.933,
|
617 |
+
"hit@10": 0.973,
|
618 |
+
"ndcg@1": 0.714,
|
619 |
+
"ndcg@5": 0.8353398361462805,
|
620 |
+
"ndcg@10": 0.8485224436537495,
|
621 |
+
"precision@1": 0.714,
|
622 |
+
"precision@5": 0.18660000000000002,
|
623 |
+
"precision@10": 0.09730000000000001,
|
624 |
+
"recall@1": 0.714,
|
625 |
+
"recall@5": 0.933,
|
626 |
+
"recall@10": 0.973,
|
627 |
+
"f1@1": 0.714,
|
628 |
+
"f1@5": 0.311,
|
629 |
+
"f1@10": 0.17690909090909088,
|
630 |
+
"map@1": 0.714,
|
631 |
+
"map@5": 0.8023833333333333,
|
632 |
+
"map@10": 0.8079615079365078,
|
633 |
+
"mrr@1": 0.714,
|
634 |
+
"mrr@5": 0.8023833333333333,
|
635 |
+
"mrr@10": 0.8079615079365078,
|
636 |
+
"num_pred": 1000,
|
637 |
+
"num_data": 1000
|
638 |
+
},
|
639 |
+
"MSCOCO_i2t": {
|
640 |
+
"hit@1": 0.672,
|
641 |
+
"hit@5": 0.928,
|
642 |
+
"hit@10": 0.982,
|
643 |
+
"ndcg@1": 0.672,
|
644 |
+
"ndcg@5": 0.8090194149563709,
|
645 |
+
"ndcg@10": 0.8266747240244816,
|
646 |
+
"precision@1": 0.672,
|
647 |
+
"precision@5": 0.18560000000000001,
|
648 |
+
"precision@10": 0.09820000000000002,
|
649 |
+
"recall@1": 0.672,
|
650 |
+
"recall@5": 0.928,
|
651 |
+
"recall@10": 0.982,
|
652 |
+
"f1@1": 0.672,
|
653 |
+
"f1@5": 0.30933333333333335,
|
654 |
+
"f1@10": 0.1785454545454545,
|
655 |
+
"map@1": 0.672,
|
656 |
+
"map@5": 0.7693666666666665,
|
657 |
+
"map@10": 0.7767690476190475,
|
658 |
+
"mrr@1": 0.672,
|
659 |
+
"mrr@5": 0.7693666666666665,
|
660 |
+
"mrr@10": 0.7767690476190475,
|
661 |
+
"num_pred": 1000,
|
662 |
+
"num_data": 1000
|
663 |
+
},
|
664 |
+
"NIGHTS": {
|
665 |
+
"hit@1": 0.668,
|
666 |
+
"hit@5": 0.978,
|
667 |
+
"hit@10": 0.993,
|
668 |
+
"ndcg@1": 0.668,
|
669 |
+
"ndcg@5": 0.8447366082363483,
|
670 |
+
"ndcg@10": 0.8496486525328769,
|
671 |
+
"precision@1": 0.668,
|
672 |
+
"precision@5": 0.19560000000000002,
|
673 |
+
"precision@10": 0.09930000000000001,
|
674 |
+
"recall@1": 0.668,
|
675 |
+
"recall@5": 0.978,
|
676 |
+
"recall@10": 0.993,
|
677 |
+
"f1@1": 0.668,
|
678 |
+
"f1@5": 0.326,
|
679 |
+
"f1@10": 0.1805454545454545,
|
680 |
+
"map@1": 0.668,
|
681 |
+
"map@5": 0.7992999999999999,
|
682 |
+
"map@10": 0.8013611111111111,
|
683 |
+
"mrr@1": 0.668,
|
684 |
+
"mrr@5": 0.7992999999999999,
|
685 |
+
"mrr@10": 0.8013611111111111,
|
686 |
+
"num_pred": 1000,
|
687 |
+
"num_data": 1000
|
688 |
+
},
|
689 |
+
"WebQA": {
|
690 |
+
"hit@1": 0.886,
|
691 |
+
"hit@5": 0.983,
|
692 |
+
"hit@10": 0.992,
|
693 |
+
"ndcg@1": 0.886,
|
694 |
+
"ndcg@5": 0.9421519699900475,
|
695 |
+
"ndcg@10": 0.9449297295649959,
|
696 |
+
"precision@1": 0.886,
|
697 |
+
"precision@5": 0.19660000000000002,
|
698 |
+
"precision@10": 0.09920000000000001,
|
699 |
+
"recall@1": 0.886,
|
700 |
+
"recall@5": 0.983,
|
701 |
+
"recall@10": 0.992,
|
702 |
+
"f1@1": 0.886,
|
703 |
+
"f1@5": 0.3276666666666666,
|
704 |
+
"f1@10": 0.18036363636363634,
|
705 |
+
"map@1": 0.886,
|
706 |
+
"map@5": 0.9281666666666666,
|
707 |
+
"map@10": 0.9292373015873016,
|
708 |
+
"mrr@1": 0.886,
|
709 |
+
"mrr@5": 0.9281666666666666,
|
710 |
+
"mrr@10": 0.9292373015873016,
|
711 |
+
"num_pred": 1000,
|
712 |
+
"num_data": 1000
|
713 |
+
},
|
714 |
+
"FashionIQ": {
|
715 |
+
"hit@1": 0.193,
|
716 |
+
"hit@5": 0.414,
|
717 |
+
"hit@10": 0.519,
|
718 |
+
"ndcg@1": 0.193,
|
719 |
+
"ndcg@5": 0.3095019379907809,
|
720 |
+
"ndcg@10": 0.3436979493850471,
|
721 |
+
"precision@1": 0.193,
|
722 |
+
"precision@5": 0.0828,
|
723 |
+
"precision@10": 0.0519,
|
724 |
+
"recall@1": 0.193,
|
725 |
+
"recall@5": 0.414,
|
726 |
+
"recall@10": 0.519,
|
727 |
+
"f1@1": 0.193,
|
728 |
+
"f1@5": 0.13800000000000004,
|
729 |
+
"f1@10": 0.09436363636363637,
|
730 |
+
"map@1": 0.193,
|
731 |
+
"map@5": 0.27481666666666665,
|
732 |
+
"map@10": 0.2890619047619048,
|
733 |
+
"mrr@1": 0.193,
|
734 |
+
"mrr@5": 0.27481666666666665,
|
735 |
+
"mrr@10": 0.2890619047619048,
|
736 |
+
"num_pred": 1000,
|
737 |
+
"num_data": 1000
|
738 |
+
},
|
739 |
+
"Wiki-SS-NQ": {
|
740 |
+
"hit@1": 0.649,
|
741 |
+
"hit@5": 0.858,
|
742 |
+
"hit@10": 0.894,
|
743 |
+
"ndcg@1": 0.649,
|
744 |
+
"ndcg@5": 0.7621993288618629,
|
745 |
+
"ndcg@10": 0.7740234309698544,
|
746 |
+
"precision@1": 0.649,
|
747 |
+
"precision@5": 0.17160000000000003,
|
748 |
+
"precision@10": 0.08940000000000001,
|
749 |
+
"recall@1": 0.649,
|
750 |
+
"recall@5": 0.858,
|
751 |
+
"recall@10": 0.894,
|
752 |
+
"f1@1": 0.649,
|
753 |
+
"f1@5": 0.286,
|
754 |
+
"f1@10": 0.16254545454545455,
|
755 |
+
"map@1": 0.649,
|
756 |
+
"map@5": 0.7301,
|
757 |
+
"map@10": 0.7350916666666667,
|
758 |
+
"mrr@1": 0.649,
|
759 |
+
"mrr@5": 0.7301,
|
760 |
+
"mrr@10": 0.7350916666666667,
|
761 |
+
"num_pred": 1000,
|
762 |
+
"num_data": 1000
|
763 |
+
},
|
764 |
+
"OVEN": {
|
765 |
+
"hit@1": 0.656,
|
766 |
+
"hit@5": 0.858,
|
767 |
+
"hit@10": 0.904,
|
768 |
+
"ndcg@1": 0.656,
|
769 |
+
"ndcg@5": 0.7688225950844196,
|
770 |
+
"ndcg@10": 0.7840840724207487,
|
771 |
+
"precision@1": 0.656,
|
772 |
+
"precision@5": 0.17160000000000003,
|
773 |
+
"precision@10": 0.09040000000000001,
|
774 |
+
"recall@1": 0.656,
|
775 |
+
"recall@5": 0.858,
|
776 |
+
"recall@10": 0.904,
|
777 |
+
"f1@1": 0.656,
|
778 |
+
"f1@5": 0.286,
|
779 |
+
"f1@10": 0.16436363636363638,
|
780 |
+
"map@1": 0.656,
|
781 |
+
"map@5": 0.7386666666666666,
|
782 |
+
"map@10": 0.7451972222222221,
|
783 |
+
"mrr@1": 0.656,
|
784 |
+
"mrr@5": 0.7386666666666666,
|
785 |
+
"mrr@10": 0.7451972222222221,
|
786 |
+
"num_pred": 1000,
|
787 |
+
"num_data": 1000
|
788 |
+
},
|
789 |
+
"EDIS": {
|
790 |
+
"hit@1": 0.81,
|
791 |
+
"hit@5": 0.958,
|
792 |
+
"hit@10": 0.985,
|
793 |
+
"ndcg@1": 0.81,
|
794 |
+
"ndcg@5": 0.8927260422814499,
|
795 |
+
"ndcg@10": 0.9013786033917063,
|
796 |
+
"precision@1": 0.81,
|
797 |
+
"precision@5": 0.19160000000000002,
|
798 |
+
"precision@10": 0.09850000000000003,
|
799 |
+
"recall@1": 0.81,
|
800 |
+
"recall@5": 0.958,
|
801 |
+
"recall@10": 0.985,
|
802 |
+
"f1@1": 0.81,
|
803 |
+
"f1@5": 0.31933333333333336,
|
804 |
+
"f1@10": 0.17909090909090908,
|
805 |
+
"map@1": 0.81,
|
806 |
+
"map@5": 0.8706833333333334,
|
807 |
+
"map@10": 0.8742063492063492,
|
808 |
+
"mrr@1": 0.81,
|
809 |
+
"mrr@5": 0.8706833333333334,
|
810 |
+
"mrr@10": 0.8742063492063492,
|
811 |
+
"num_pred": 1000,
|
812 |
+
"num_data": 1000
|
813 |
+
},
|
814 |
+
"MSCOCO": {
|
815 |
+
"hit@1": 0.616,
|
816 |
+
"hit@5": 0.779,
|
817 |
+
"hit@10": 0.832,
|
818 |
+
"ndcg@1": 0.616,
|
819 |
+
"ndcg@5": 0.705461859923422,
|
820 |
+
"ndcg@10": 0.7225744447775342,
|
821 |
+
"precision@1": 0.616,
|
822 |
+
"precision@5": 0.15580000000000002,
|
823 |
+
"precision@10": 0.08320000000000002,
|
824 |
+
"recall@1": 0.616,
|
825 |
+
"recall@5": 0.779,
|
826 |
+
"recall@10": 0.832,
|
827 |
+
"f1@1": 0.616,
|
828 |
+
"f1@5": 0.25966666666666677,
|
829 |
+
"f1@10": 0.15127272727272725,
|
830 |
+
"map@1": 0.616,
|
831 |
+
"map@5": 0.6808166666666666,
|
832 |
+
"map@10": 0.6878607142857143,
|
833 |
+
"mrr@1": 0.616,
|
834 |
+
"mrr@5": 0.6808166666666666,
|
835 |
+
"mrr@10": 0.6878607142857143,
|
836 |
+
"num_pred": 1000,
|
837 |
+
"num_data": 1000
|
838 |
+
},
|
839 |
+
"RefCOCO": {
|
840 |
+
"hit@1": 0.83,
|
841 |
+
"hit@5": 0.965,
|
842 |
+
"hit@10": 0.982,
|
843 |
+
"ndcg@1": 0.83,
|
844 |
+
"ndcg@5": 0.907274694038521,
|
845 |
+
"ndcg@10": 0.9127997729844821,
|
846 |
+
"precision@1": 0.83,
|
847 |
+
"precision@5": 0.19300000000000003,
|
848 |
+
"precision@10": 0.09820000000000002,
|
849 |
+
"recall@1": 0.83,
|
850 |
+
"recall@5": 0.965,
|
851 |
+
"recall@10": 0.982,
|
852 |
+
"f1@1": 0.83,
|
853 |
+
"f1@5": 0.3216666666666667,
|
854 |
+
"f1@10": 0.17854545454545453,
|
855 |
+
"map@1": 0.83,
|
856 |
+
"map@5": 0.8875666666666666,
|
857 |
+
"map@10": 0.8898630952380951,
|
858 |
+
"mrr@1": 0.83,
|
859 |
+
"mrr@5": 0.8875666666666666,
|
860 |
+
"mrr@10": 0.8898630952380951,
|
861 |
+
"num_pred": 1000,
|
862 |
+
"num_data": 1000
|
863 |
+
},
|
864 |
+
"RefCOCO-Matching": {
|
865 |
+
"hit@1": 0.85,
|
866 |
+
"hit@5": 0.999,
|
867 |
+
"hit@10": 1.0,
|
868 |
+
"ndcg@1": 0.85,
|
869 |
+
"ndcg@5": 0.9446891964404038,
|
870 |
+
"ndcg@10": 0.9450046613171896,
|
871 |
+
"precision@1": 0.85,
|
872 |
+
"precision@5": 0.19980000000000003,
|
873 |
+
"precision@10": 0.10000000000000002,
|
874 |
+
"recall@1": 0.85,
|
875 |
+
"recall@5": 0.999,
|
876 |
+
"recall@10": 1.0,
|
877 |
+
"f1@1": 0.85,
|
878 |
+
"f1@5": 0.333,
|
879 |
+
"f1@10": 0.18181818181818182,
|
880 |
+
"map@1": 0.85,
|
881 |
+
"map@5": 0.9259833333333334,
|
882 |
+
"map@10": 0.9261083333333333,
|
883 |
+
"mrr@1": 0.85,
|
884 |
+
"mrr@5": 0.9229833333333334,
|
885 |
+
"mrr@10": 0.9231083333333333,
|
886 |
+
"num_pred": 1000,
|
887 |
+
"num_data": 1000
|
888 |
+
},
|
889 |
+
"Visual7W-Pointing": {
|
890 |
+
"hit@1": 0.71,
|
891 |
+
"hit@5": 0.871,
|
892 |
+
"hit@10": 0.909,
|
893 |
+
"ndcg@1": 0.71,
|
894 |
+
"ndcg@5": 0.7972840336402242,
|
895 |
+
"ndcg@10": 0.8094017818760351,
|
896 |
+
"precision@1": 0.71,
|
897 |
+
"precision@5": 0.17420000000000002,
|
898 |
+
"precision@10": 0.09090000000000001,
|
899 |
+
"recall@1": 0.71,
|
900 |
+
"recall@5": 0.871,
|
901 |
+
"recall@10": 0.909,
|
902 |
+
"f1@1": 0.71,
|
903 |
+
"f1@5": 0.2903333333333334,
|
904 |
+
"f1@10": 0.16527272727272727,
|
905 |
+
"map@1": 0.71,
|
906 |
+
"map@5": 0.7726333333333332,
|
907 |
+
"map@10": 0.7775396825396825,
|
908 |
+
"mrr@1": 0.71,
|
909 |
+
"mrr@5": 0.7726333333333332,
|
910 |
+
"mrr@10": 0.7775396825396825,
|
911 |
+
"num_pred": 1000,
|
912 |
+
"num_data": 1000
|
913 |
+
}
|
914 |
+
},
|
915 |
+
"visdoc": {
|
916 |
+
"ViDoRe_arxivqa": {
|
917 |
+
"hit@1": 0.652,
|
918 |
+
"hit@5": 0.808,
|
919 |
+
"hit@10": 0.856,
|
920 |
+
"ndcg@1": 0.652,
|
921 |
+
"ndcg@5": 0.7388343621431214,
|
922 |
+
"ndcg@10": 0.7543346522533037,
|
923 |
+
"precision@1": 0.652,
|
924 |
+
"precision@5": 0.1616,
|
925 |
+
"precision@10": 0.0856,
|
926 |
+
"recall@1": 0.652,
|
927 |
+
"recall@5": 0.808,
|
928 |
+
"recall@10": 0.856,
|
929 |
+
"f1@1": 0.652,
|
930 |
+
"f1@5": 0.26933333333333337,
|
931 |
+
"f1@10": 0.15563636363636363,
|
932 |
+
"map@1": 0.652,
|
933 |
+
"map@5": 0.7155,
|
934 |
+
"map@10": 0.7218817460317459,
|
935 |
+
"mrr@1": 0.652,
|
936 |
+
"mrr@5": 0.7155,
|
937 |
+
"mrr@10": 0.7218817460317459,
|
938 |
+
"num_pred": 500,
|
939 |
+
"num_data": 500
|
940 |
+
},
|
941 |
+
"ViDoRe_docvqa": {
|
942 |
+
"hit@1": 0.30155210643015523,
|
943 |
+
"hit@5": 0.49889135254988914,
|
944 |
+
"hit@10": 0.5898004434589801,
|
945 |
+
"ndcg@1": 0.30155210643015523,
|
946 |
+
"ndcg@5": 0.40271180419162816,
|
947 |
+
"ndcg@10": 0.4313200960885016,
|
948 |
+
"precision@1": 0.30155210643015523,
|
949 |
+
"precision@5": 0.10022172949002216,
|
950 |
+
"precision@10": 0.05964523281596451,
|
951 |
+
"recall@1": 0.30155210643015523,
|
952 |
+
"recall@5": 0.49379157427937914,
|
953 |
+
"recall@10": 0.5826681448632668,
|
954 |
+
"f1@1": 0.30155210643015523,
|
955 |
+
"f1@5": 0.16621264913947845,
|
956 |
+
"f1@10": 0.10783444198078346,
|
957 |
+
"map@1": 0.30155210643015523,
|
958 |
+
"map@5": 0.3719327420546933,
|
959 |
+
"map@10": 0.38349215147995636,
|
960 |
+
"mrr@1": 0.30155210643015523,
|
961 |
+
"mrr@5": 0.3732446415373244,
|
962 |
+
"mrr@10": 0.38493031358885016,
|
963 |
+
"num_pred": 451,
|
964 |
+
"num_data": 451
|
965 |
+
},
|
966 |
+
"ViDoRe_infovqa": {
|
967 |
+
"hit@1": 0.6740890688259109,
|
968 |
+
"hit@5": 0.8663967611336032,
|
969 |
+
"hit@10": 0.9048582995951417,
|
970 |
+
"ndcg@1": 0.6740890688259109,
|
971 |
+
"ndcg@5": 0.7801033100100705,
|
972 |
+
"ndcg@10": 0.7930116387134283,
|
973 |
+
"precision@1": 0.6740890688259109,
|
974 |
+
"precision@5": 0.17327935222672064,
|
975 |
+
"precision@10": 0.09068825910931173,
|
976 |
+
"recall@1": 0.6740890688259109,
|
977 |
+
"recall@5": 0.8653846153846154,
|
978 |
+
"recall@10": 0.9048582995951417,
|
979 |
+
"f1@1": 0.6740890688259109,
|
980 |
+
"f1@5": 0.2887025255446309,
|
981 |
+
"f1@10": 0.16482640166850696,
|
982 |
+
"map@1": 0.6740890688259109,
|
983 |
+
"map@5": 0.7511470985155195,
|
984 |
+
"map@10": 0.7566167020114387,
|
985 |
+
"mrr@1": 0.6740890688259109,
|
986 |
+
"mrr@5": 0.7516531713900135,
|
987 |
+
"mrr@10": 0.7568335903862219,
|
988 |
+
"num_pred": 494,
|
989 |
+
"num_data": 494
|
990 |
+
},
|
991 |
+
"ViDoRe_tabfquad": {
|
992 |
+
"hit@1": 0.7107142857142857,
|
993 |
+
"hit@5": 0.8964285714285715,
|
994 |
+
"hit@10": 0.9464285714285714,
|
995 |
+
"ndcg@1": 0.7107142857142857,
|
996 |
+
"ndcg@5": 0.8138755399252038,
|
997 |
+
"ndcg@10": 0.8304008505590015,
|
998 |
+
"precision@1": 0.7107142857142857,
|
999 |
+
"precision@5": 0.17928571428571427,
|
1000 |
+
"precision@10": 0.09464285714285714,
|
1001 |
+
"recall@1": 0.7107142857142857,
|
1002 |
+
"recall@5": 0.8964285714285715,
|
1003 |
+
"recall@10": 0.9464285714285714,
|
1004 |
+
"f1@1": 0.7107142857142857,
|
1005 |
+
"f1@5": 0.29880952380952386,
|
1006 |
+
"f1@10": 0.1720779220779221,
|
1007 |
+
"map@1": 0.7107142857142857,
|
1008 |
+
"map@5": 0.785952380952381,
|
1009 |
+
"map@10": 0.7929747732426303,
|
1010 |
+
"mrr@1": 0.7107142857142857,
|
1011 |
+
"mrr@5": 0.785952380952381,
|
1012 |
+
"mrr@10": 0.7929747732426303,
|
1013 |
+
"num_pred": 280,
|
1014 |
+
"num_data": 280
|
1015 |
+
},
|
1016 |
+
"ViDoRe_tatdqa": {
|
1017 |
+
"hit@1": 0.24665856622114216,
|
1018 |
+
"hit@5": 0.48541919805589306,
|
1019 |
+
"hit@10": 0.6026731470230863,
|
1020 |
+
"ndcg@1": 0.24665856622114216,
|
1021 |
+
"ndcg@5": 0.3736179555419941,
|
1022 |
+
"ndcg@10": 0.4114961956617735,
|
1023 |
+
"precision@1": 0.24665856622114216,
|
1024 |
+
"precision@5": 0.09744835965978127,
|
1025 |
+
"precision@10": 0.06051032806804374,
|
1026 |
+
"recall@1": 0.24605103280680438,
|
1027 |
+
"recall@5": 0.4848116646415553,
|
1028 |
+
"recall@10": 0.6023693803159174,
|
1029 |
+
"f1@1": 0.24625354394491697,
|
1030 |
+
"f1@5": 0.16218249146560204,
|
1031 |
+
"f1@10": 0.10993593283994259,
|
1032 |
+
"map@1": 0.24665856622114216,
|
1033 |
+
"map@5": 0.3365127582017011,
|
1034 |
+
"map@10": 0.3520810430287951,
|
1035 |
+
"mrr@1": 0.24665856622114216,
|
1036 |
+
"mrr@5": 0.33676589712434185,
|
1037 |
+
"mrr@10": 0.35225824027464364,
|
1038 |
+
"num_pred": 1646,
|
1039 |
+
"num_data": 1646
|
1040 |
+
},
|
1041 |
+
"ViDoRe_shiftproject": {
|
1042 |
+
"hit@1": 0.41,
|
1043 |
+
"hit@5": 0.77,
|
1044 |
+
"hit@10": 0.85,
|
1045 |
+
"ndcg@1": 0.41,
|
1046 |
+
"ndcg@5": 0.5978201649268686,
|
1047 |
+
"ndcg@10": 0.6247804161088925,
|
1048 |
+
"precision@1": 0.41,
|
1049 |
+
"precision@5": 0.15399999999999997,
|
1050 |
+
"precision@10": 0.08499999999999996,
|
1051 |
+
"recall@1": 0.41,
|
1052 |
+
"recall@5": 0.77,
|
1053 |
+
"recall@10": 0.85,
|
1054 |
+
"f1@1": 0.41,
|
1055 |
+
"f1@5": 0.2566666666666667,
|
1056 |
+
"f1@10": 0.15454545454545457,
|
1057 |
+
"map@1": 0.41,
|
1058 |
+
"map@5": 0.5408333333333333,
|
1059 |
+
"map@10": 0.552607142857143,
|
1060 |
+
"mrr@1": 0.41,
|
1061 |
+
"mrr@5": 0.5408333333333333,
|
1062 |
+
"mrr@10": 0.552607142857143,
|
1063 |
+
"num_pred": 100,
|
1064 |
+
"num_data": 100
|
1065 |
+
},
|
1066 |
+
"ViDoRe_syntheticDocQA_artificial_intelligence": {
|
1067 |
+
"hit@1": 0.64,
|
1068 |
+
"hit@5": 0.89,
|
1069 |
+
"hit@10": 0.89,
|
1070 |
+
"ndcg@1": 0.64,
|
1071 |
+
"ndcg@5": 0.7720316911918113,
|
1072 |
+
"ndcg@10": 0.7720316911918113,
|
1073 |
+
"precision@1": 0.64,
|
1074 |
+
"precision@5": 0.17799999999999996,
|
1075 |
+
"precision@10": 0.08899999999999998,
|
1076 |
+
"recall@1": 0.64,
|
1077 |
+
"recall@5": 0.89,
|
1078 |
+
"recall@10": 0.89,
|
1079 |
+
"f1@1": 0.64,
|
1080 |
+
"f1@5": 0.2966666666666667,
|
1081 |
+
"f1@10": 0.16181818181818186,
|
1082 |
+
"map@1": 0.64,
|
1083 |
+
"map@5": 0.7328333333333333,
|
1084 |
+
"map@10": 0.7328333333333333,
|
1085 |
+
"mrr@1": 0.64,
|
1086 |
+
"mrr@5": 0.7328333333333333,
|
1087 |
+
"mrr@10": 0.7328333333333333,
|
1088 |
+
"num_pred": 100,
|
1089 |
+
"num_data": 100
|
1090 |
+
},
|
1091 |
+
"ViDoRe_syntheticDocQA_energy": {
|
1092 |
+
"hit@1": 0.66,
|
1093 |
+
"hit@5": 0.86,
|
1094 |
+
"hit@10": 0.91,
|
1095 |
+
"ndcg@1": 0.66,
|
1096 |
+
"ndcg@5": 0.7686731569412208,
|
1097 |
+
"ndcg@10": 0.7839527355254091,
|
1098 |
+
"precision@1": 0.66,
|
1099 |
+
"precision@5": 0.172,
|
1100 |
+
"precision@10": 0.09099999999999998,
|
1101 |
+
"recall@1": 0.66,
|
1102 |
+
"recall@5": 0.86,
|
1103 |
+
"recall@10": 0.91,
|
1104 |
+
"f1@1": 0.66,
|
1105 |
+
"f1@5": 0.28666666666666674,
|
1106 |
+
"f1@10": 0.16545454545454555,
|
1107 |
+
"map@1": 0.66,
|
1108 |
+
"map@5": 0.7381666666666666,
|
1109 |
+
"map@10": 0.7439563492063493,
|
1110 |
+
"mrr@1": 0.66,
|
1111 |
+
"mrr@5": 0.7381666666666666,
|
1112 |
+
"mrr@10": 0.7439563492063493,
|
1113 |
+
"num_pred": 100,
|
1114 |
+
"num_data": 100
|
1115 |
+
},
|
1116 |
+
"ViDoRe_syntheticDocQA_government_reports": {
|
1117 |
+
"hit@1": 0.59,
|
1118 |
+
"hit@5": 0.91,
|
1119 |
+
"hit@10": 0.95,
|
1120 |
+
"ndcg@1": 0.59,
|
1121 |
+
"ndcg@5": 0.7654800565201222,
|
1122 |
+
"ndcg@10": 0.7789478335522557,
|
1123 |
+
"precision@1": 0.59,
|
1124 |
+
"precision@5": 0.182,
|
1125 |
+
"precision@10": 0.09499999999999999,
|
1126 |
+
"recall@1": 0.59,
|
1127 |
+
"recall@5": 0.91,
|
1128 |
+
"recall@10": 0.95,
|
1129 |
+
"f1@1": 0.59,
|
1130 |
+
"f1@5": 0.3033333333333334,
|
1131 |
+
"f1@10": 0.1727272727272728,
|
1132 |
+
"map@1": 0.59,
|
1133 |
+
"map@5": 0.7166666666666666,
|
1134 |
+
"map@10": 0.7225396825396825,
|
1135 |
+
"mrr@1": 0.59,
|
1136 |
+
"mrr@5": 0.7166666666666666,
|
1137 |
+
"mrr@10": 0.7225396825396825,
|
1138 |
+
"num_pred": 100,
|
1139 |
+
"num_data": 100
|
1140 |
+
},
|
1141 |
+
"ViDoRe_syntheticDocQA_healthcare_industry": {
|
1142 |
+
"hit@1": 0.71,
|
1143 |
+
"hit@5": 0.9,
|
1144 |
+
"hit@10": 0.94,
|
1145 |
+
"ndcg@1": 0.71,
|
1146 |
+
"ndcg@5": 0.8075594917815855,
|
1147 |
+
"ndcg@10": 0.8212560073514661,
|
1148 |
+
"precision@1": 0.71,
|
1149 |
+
"precision@5": 0.17999999999999997,
|
1150 |
+
"precision@10": 0.09399999999999999,
|
1151 |
+
"recall@1": 0.71,
|
1152 |
+
"recall@5": 0.9,
|
1153 |
+
"recall@10": 0.94,
|
1154 |
+
"f1@1": 0.71,
|
1155 |
+
"f1@5": 0.30000000000000004,
|
1156 |
+
"f1@10": 0.17090909090909098,
|
1157 |
+
"map@1": 0.71,
|
1158 |
+
"map@5": 0.7771666666666667,
|
1159 |
+
"map@10": 0.7832777777777777,
|
1160 |
+
"mrr@1": 0.71,
|
1161 |
+
"mrr@5": 0.7771666666666667,
|
1162 |
+
"mrr@10": 0.7832777777777777,
|
1163 |
+
"num_pred": 100,
|
1164 |
+
"num_data": 100
|
1165 |
+
},
|
1166 |
+
"VisRAG_ArxivQA": {
|
1167 |
+
"hit@1": 0.5514705882352942,
|
1168 |
+
"hit@5": 0.696078431372549,
|
1169 |
+
"hit@10": 0.7475490196078431,
|
1170 |
+
"ndcg@1": 0.5514705882352942,
|
1171 |
+
"ndcg@5": 0.6310013830686256,
|
1172 |
+
"ndcg@10": 0.6475995079802747,
|
1173 |
+
"precision@1": 0.5514705882352942,
|
1174 |
+
"precision@5": 0.1392156862745098,
|
1175 |
+
"precision@10": 0.07475490196078431,
|
1176 |
+
"recall@1": 0.5514705882352942,
|
1177 |
+
"recall@5": 0.696078431372549,
|
1178 |
+
"recall@10": 0.7475490196078431,
|
1179 |
+
"f1@1": 0.5514705882352942,
|
1180 |
+
"f1@5": 0.2320261437908497,
|
1181 |
+
"f1@10": 0.13591800356506242,
|
1182 |
+
"map@1": 0.5514705882352942,
|
1183 |
+
"map@5": 0.6091503267973857,
|
1184 |
+
"map@10": 0.6159649276377217,
|
1185 |
+
"mrr@1": 0.5514705882352942,
|
1186 |
+
"mrr@5": 0.6091503267973857,
|
1187 |
+
"mrr@10": 0.6159649276377217,
|
1188 |
+
"num_pred": 816,
|
1189 |
+
"num_data": 816
|
1190 |
+
},
|
1191 |
+
"VisRAG_ChartQA": {
|
1192 |
+
"hit@1": 0.7619047619047619,
|
1193 |
+
"hit@5": 0.8571428571428571,
|
1194 |
+
"hit@10": 0.9047619047619048,
|
1195 |
+
"ndcg@1": 0.7619047619047619,
|
1196 |
+
"ndcg@5": 0.8178368097505687,
|
1197 |
+
"ndcg@10": 0.83443597971167,
|
1198 |
+
"precision@1": 0.7619047619047619,
|
1199 |
+
"precision@5": 0.17142857142857135,
|
1200 |
+
"precision@10": 0.09047619047619043,
|
1201 |
+
"recall@1": 0.7619047619047619,
|
1202 |
+
"recall@5": 0.8571428571428571,
|
1203 |
+
"recall@10": 0.9047619047619048,
|
1204 |
+
"f1@1": 0.7619047619047619,
|
1205 |
+
"f1@5": 0.28571428571428564,
|
1206 |
+
"f1@10": 0.16450216450216454,
|
1207 |
+
"map@1": 0.7619047619047619,
|
1208 |
+
"map@5": 0.8042328042328043,
|
1209 |
+
"map@10": 0.8117913832199546,
|
1210 |
+
"mrr@1": 0.7619047619047619,
|
1211 |
+
"mrr@5": 0.8042328042328043,
|
1212 |
+
"mrr@10": 0.8117913832199546,
|
1213 |
+
"num_pred": 63,
|
1214 |
+
"num_data": 63
|
1215 |
+
},
|
1216 |
+
"VisRAG_MP-DocVQA": {
|
1217 |
+
"hit@1": 0.544839255499154,
|
1218 |
+
"hit@5": 0.7495769881556683,
|
1219 |
+
"hit@10": 0.7969543147208121,
|
1220 |
+
"ndcg@1": 0.544839255499154,
|
1221 |
+
"ndcg@5": 0.6536748593721804,
|
1222 |
+
"ndcg@10": 0.6687163141394652,
|
1223 |
+
"precision@1": 0.544839255499154,
|
1224 |
+
"precision@5": 0.14991539763113365,
|
1225 |
+
"precision@10": 0.0796954314720812,
|
1226 |
+
"recall@1": 0.544839255499154,
|
1227 |
+
"recall@5": 0.7495769881556683,
|
1228 |
+
"recall@10": 0.7969543147208121,
|
1229 |
+
"f1@1": 0.544839255499154,
|
1230 |
+
"f1@5": 0.24985899605188952,
|
1231 |
+
"f1@10": 0.14490078449469315,
|
1232 |
+
"map@1": 0.544839255499154,
|
1233 |
+
"map@5": 0.6217992103778907,
|
1234 |
+
"map@10": 0.6278469368570891,
|
1235 |
+
"mrr@1": 0.544839255499154,
|
1236 |
+
"mrr@5": 0.6217992103778907,
|
1237 |
+
"mrr@10": 0.6278469368570891,
|
1238 |
+
"num_pred": 591,
|
1239 |
+
"num_data": 591
|
1240 |
+
},
|
1241 |
+
"VisRAG_SlideVQA": {
|
1242 |
+
"hit@1": 0.7320143884892086,
|
1243 |
+
"hit@5": 0.8992805755395683,
|
1244 |
+
"hit@10": 0.9370503597122302,
|
1245 |
+
"ndcg@1": 0.7320143884892086,
|
1246 |
+
"ndcg@5": 0.8009145701404112,
|
1247 |
+
"ndcg@10": 0.8169189758450583,
|
1248 |
+
"precision@1": 0.7320143884892086,
|
1249 |
+
"precision@5": 0.2183453237410072,
|
1250 |
+
"precision@10": 0.11510791366906474,
|
1251 |
+
"recall@1": 0.625,
|
1252 |
+
"recall@5": 0.8714028776978417,
|
1253 |
+
"recall@10": 0.9172661870503597,
|
1254 |
+
"f1@1": 0.6606714628297362,
|
1255 |
+
"f1@5": 0.3429256594724221,
|
1256 |
+
"f1@10": 0.20220187486374538,
|
1257 |
+
"map@1": 0.7320143884892086,
|
1258 |
+
"map@5": 0.7641037170263789,
|
1259 |
+
"map@10": 0.7719267443188306,
|
1260 |
+
"mrr@1": 0.7320143884892086,
|
1261 |
+
"mrr@5": 0.8017086330935251,
|
1262 |
+
"mrr@10": 0.8068252540824483,
|
1263 |
+
"num_pred": 556,
|
1264 |
+
"num_data": 556
|
1265 |
+
},
|
1266 |
+
"VisRAG_InfoVQA": {
|
1267 |
+
"hit@1": 0.6267409470752089,
|
1268 |
+
"hit@5": 0.871866295264624,
|
1269 |
+
"hit@10": 0.9289693593314763,
|
1270 |
+
"ndcg@1": 0.6267409470752089,
|
1271 |
+
"ndcg@5": 0.7597297179067588,
|
1272 |
+
"ndcg@10": 0.7785227500535231,
|
1273 |
+
"precision@1": 0.6267409470752089,
|
1274 |
+
"precision@5": 0.17437325905292478,
|
1275 |
+
"precision@10": 0.09289693593314761,
|
1276 |
+
"recall@1": 0.6267409470752089,
|
1277 |
+
"recall@5": 0.871866295264624,
|
1278 |
+
"recall@10": 0.9289693593314763,
|
1279 |
+
"f1@1": 0.6267409470752089,
|
1280 |
+
"f1@5": 0.2906220984215414,
|
1281 |
+
"f1@10": 0.1689035198784503,
|
1282 |
+
"map@1": 0.6267409470752089,
|
1283 |
+
"map@5": 0.7221680594243268,
|
1284 |
+
"map@10": 0.7301156209930583,
|
1285 |
+
"mrr@1": 0.6267409470752089,
|
1286 |
+
"mrr@5": 0.7221680594243268,
|
1287 |
+
"mrr@10": 0.7301156209930583,
|
1288 |
+
"num_pred": 718,
|
1289 |
+
"num_data": 718
|
1290 |
+
},
|
1291 |
+
"VisRAG_PlotQA": {
|
1292 |
+
"hit@1": 0.37891077636152953,
|
1293 |
+
"hit@5": 0.6500579374275782,
|
1294 |
+
"hit@10": 0.7531865585168018,
|
1295 |
+
"ndcg@1": 0.37891077636152953,
|
1296 |
+
"ndcg@5": 0.5200085565479549,
|
1297 |
+
"ndcg@10": 0.5537339394426217,
|
1298 |
+
"precision@1": 0.37891077636152953,
|
1299 |
+
"precision@5": 0.13001158748551564,
|
1300 |
+
"precision@10": 0.07531865585168018,
|
1301 |
+
"recall@1": 0.37891077636152953,
|
1302 |
+
"recall@5": 0.6500579374275782,
|
1303 |
+
"recall@10": 0.7531865585168018,
|
1304 |
+
"f1@1": 0.37891077636152953,
|
1305 |
+
"f1@5": 0.21668597914252613,
|
1306 |
+
"f1@10": 0.13694301063941855,
|
1307 |
+
"map@1": 0.37891077636152953,
|
1308 |
+
"map@5": 0.4770760911548861,
|
1309 |
+
"map@10": 0.4912132833048244,
|
1310 |
+
"mrr@1": 0.37891077636152953,
|
1311 |
+
"mrr@5": 0.4770760911548861,
|
1312 |
+
"mrr@10": 0.4912132833048244,
|
1313 |
+
"num_pred": 863,
|
1314 |
+
"num_data": 863
|
1315 |
+
},
|
1316 |
+
"ViDoSeek-page": "FILE_N/A",
|
1317 |
+
"ViDoSeek-doc": "FILE_N/A",
|
1318 |
+
"MMLongBench-page": "FILE_N/A",
|
1319 |
+
"MMLongBench-doc": "FILE_N/A"
|
1320 |
+
},
|
1321 |
+
"video": {
|
1322 |
+
"DiDeMo": {
|
1323 |
+
"hit@1": 0.41434262948207173,
|
1324 |
+
"hit@5": 0.6673306772908366,
|
1325 |
+
"hit@10": 0.7669322709163346,
|
1326 |
+
"ndcg@1": 0.41434262948207173,
|
1327 |
+
"ndcg@5": 0.5482214829308661,
|
1328 |
+
"ndcg@10": 0.5804168282669133,
|
1329 |
+
"precision@1": 0.41434262948207173,
|
1330 |
+
"precision@5": 0.13346613545816732,
|
1331 |
+
"precision@10": 0.07669322709163347,
|
1332 |
+
"recall@1": 0.41434262948207173,
|
1333 |
+
"recall@5": 0.6673306772908366,
|
1334 |
+
"recall@10": 0.7669322709163346,
|
1335 |
+
"f1@1": 0.41434262948207173,
|
1336 |
+
"f1@5": 0.22244355909694558,
|
1337 |
+
"f1@10": 0.1394422310756972,
|
1338 |
+
"map@1": 0.41434262948207173,
|
1339 |
+
"map@5": 0.5087317397078354,
|
1340 |
+
"map@10": 0.5220012805919181,
|
1341 |
+
"mrr@1": 0.41434262948207173,
|
1342 |
+
"mrr@5": 0.5087317397078354,
|
1343 |
+
"mrr@10": 0.5220012805919181,
|
1344 |
+
"num_pred": 1004,
|
1345 |
+
"num_data": 1004
|
1346 |
+
},
|
1347 |
+
"MSR-VTT": {
|
1348 |
+
"hit@1": 0.395,
|
1349 |
+
"hit@5": 0.651,
|
1350 |
+
"hit@10": 0.743,
|
1351 |
+
"ndcg@1": 0.395,
|
1352 |
+
"ndcg@5": 0.5315341508628847,
|
1353 |
+
"ndcg@10": 0.5613539463855975,
|
1354 |
+
"precision@1": 0.395,
|
1355 |
+
"precision@5": 0.13019999999999998,
|
1356 |
+
"precision@10": 0.07429999999999999,
|
1357 |
+
"recall@1": 0.395,
|
1358 |
+
"recall@5": 0.651,
|
1359 |
+
"recall@10": 0.743,
|
1360 |
+
"f1@1": 0.395,
|
1361 |
+
"f1@5": 0.21700000000000005,
|
1362 |
+
"f1@10": 0.13509090909090912,
|
1363 |
+
"map@1": 0.395,
|
1364 |
+
"map@5": 0.4917833333333333,
|
1365 |
+
"map@10": 0.5041376984126984,
|
1366 |
+
"mrr@1": 0.395,
|
1367 |
+
"mrr@5": 0.4917833333333333,
|
1368 |
+
"mrr@10": 0.5041376984126984,
|
1369 |
+
"num_pred": 1000,
|
1370 |
+
"num_data": 1000
|
1371 |
+
},
|
1372 |
+
"MSVD": {
|
1373 |
+
"hit@1": 0.5701492537313433,
|
1374 |
+
"hit@5": 0.8462686567164179,
|
1375 |
+
"hit@10": 0.9208955223880597,
|
1376 |
+
"ndcg@1": 0.5701492537313433,
|
1377 |
+
"ndcg@5": 0.723657532845859,
|
1378 |
+
"ndcg@10": 0.7478301317959454,
|
1379 |
+
"precision@1": 0.5701492537313433,
|
1380 |
+
"precision@5": 0.16925373134328356,
|
1381 |
+
"precision@10": 0.09208955223880595,
|
1382 |
+
"recall@1": 0.5701492537313433,
|
1383 |
+
"recall@5": 0.8462686567164179,
|
1384 |
+
"recall@10": 0.9208955223880597,
|
1385 |
+
"f1@1": 0.5701492537313433,
|
1386 |
+
"f1@5": 0.28208955223880605,
|
1387 |
+
"f1@10": 0.16743554952510178,
|
1388 |
+
"map@1": 0.5701492537313433,
|
1389 |
+
"map@5": 0.6822636815920398,
|
1390 |
+
"map@10": 0.6922648661454632,
|
1391 |
+
"mrr@1": 0.5701492537313433,
|
1392 |
+
"mrr@5": 0.6822636815920398,
|
1393 |
+
"mrr@10": 0.6922648661454632,
|
1394 |
+
"num_pred": 670,
|
1395 |
+
"num_data": 670
|
1396 |
+
},
|
1397 |
+
"VATEX": {
|
1398 |
+
"hit@1": 0.28829834747655203,
|
1399 |
+
"hit@5": 0.550915587315766,
|
1400 |
+
"hit@10": 0.6677087985707906,
|
1401 |
+
"ndcg@1": 0.28829834747655203,
|
1402 |
+
"ndcg@5": 0.42576353345131696,
|
1403 |
+
"ndcg@10": 0.4636011840646212,
|
1404 |
+
"precision@1": 0.28829834747655203,
|
1405 |
+
"precision@5": 0.1101831174631532,
|
1406 |
+
"precision@10": 0.06677087985707905,
|
1407 |
+
"recall@1": 0.28829834747655203,
|
1408 |
+
"recall@5": 0.550915587315766,
|
1409 |
+
"recall@10": 0.6677087985707906,
|
1410 |
+
"f1@1": 0.28829834747655203,
|
1411 |
+
"f1@5": 0.18363852910525535,
|
1412 |
+
"f1@10": 0.12140159974014374,
|
1413 |
+
"map@1": 0.28829834747655203,
|
1414 |
+
"map@5": 0.3842898615453328,
|
1415 |
+
"map@10": 0.3999484251047449,
|
1416 |
+
"mrr@1": 0.28829834747655203,
|
1417 |
+
"mrr@5": 0.3842898615453328,
|
1418 |
+
"mrr@10": 0.3999484251047449,
|
1419 |
+
"num_pred": 4478,
|
1420 |
+
"num_data": 4478
|
1421 |
+
},
|
1422 |
+
"YouCook2": {
|
1423 |
+
"hit@1": 0.0994023277760302,
|
1424 |
+
"hit@5": 0.22963195973576597,
|
1425 |
+
"hit@10": 0.3057565272098144,
|
1426 |
+
"ndcg@1": 0.0994023277760302,
|
1427 |
+
"ndcg@5": 0.16505765738983919,
|
1428 |
+
"ndcg@10": 0.1896536372501136,
|
1429 |
+
"precision@1": 0.0994023277760302,
|
1430 |
+
"precision@5": 0.04592639194715319,
|
1431 |
+
"precision@10": 0.030575652720981446,
|
1432 |
+
"recall@1": 0.0994023277760302,
|
1433 |
+
"recall@5": 0.22963195973576597,
|
1434 |
+
"recall@10": 0.3057565272098144,
|
1435 |
+
"f1@1": 0.0994023277760302,
|
1436 |
+
"f1@5": 0.07654398657858867,
|
1437 |
+
"f1@10": 0.0555920958563299,
|
1438 |
+
"map@1": 0.0994023277760302,
|
1439 |
+
"map@5": 0.14391842298416693,
|
1440 |
+
"map@10": 0.15405713087373987,
|
1441 |
+
"mrr@1": 0.0994023277760302,
|
1442 |
+
"mrr@5": 0.14391842298416693,
|
1443 |
+
"mrr@10": 0.15405713087373987,
|
1444 |
+
"num_pred": 3179,
|
1445 |
+
"num_data": 3179
|
1446 |
+
},
|
1447 |
+
"QVHighlight": {
|
1448 |
+
"hit@1": 0.45891043397968606,
|
1449 |
+
"hit@5": 0.8337950138504155,
|
1450 |
+
"hit@10": 1.0,
|
1451 |
+
"ndcg@1": 0.45891043397968606,
|
1452 |
+
"ndcg@5": 0.6547709307490372,
|
1453 |
+
"ndcg@10": 0.7084382760829699,
|
1454 |
+
"precision@1": 0.45891043397968606,
|
1455 |
+
"precision@5": 0.1667590027700831,
|
1456 |
+
"precision@10": 0.09999999999999998,
|
1457 |
+
"recall@1": 0.45891043397968606,
|
1458 |
+
"recall@5": 0.8337950138504155,
|
1459 |
+
"recall@10": 1.0,
|
1460 |
+
"f1@1": 0.45891043397968606,
|
1461 |
+
"f1@5": 0.27793167128347196,
|
1462 |
+
"f1@10": 0.18181818181818185,
|
1463 |
+
"map@1": 0.45891043397968606,
|
1464 |
+
"map@5": 0.5954293628808864,
|
1465 |
+
"map@10": 0.6175409283442524,
|
1466 |
+
"mrr@1": 0.45891043397968606,
|
1467 |
+
"mrr@5": 0.5954293628808864,
|
1468 |
+
"mrr@10": 0.6175409283442524,
|
1469 |
+
"num_pred": 1083,
|
1470 |
+
"num_data": 1083
|
1471 |
+
},
|
1472 |
+
"Charades-STA": {
|
1473 |
+
"hit@1": 0.15268225584594222,
|
1474 |
+
"hit@5": 0.546079779917469,
|
1475 |
+
"hit@10": 1.0,
|
1476 |
+
"ndcg@1": 0.15268225584594222,
|
1477 |
+
"ndcg@5": 0.3513901183213191,
|
1478 |
+
"ndcg@10": 0.49393804155442406,
|
1479 |
+
"precision@1": 0.15268225584594222,
|
1480 |
+
"precision@5": 0.1092159559834938,
|
1481 |
+
"precision@10": 0.09999999999999998,
|
1482 |
+
"recall@1": 0.15268225584594222,
|
1483 |
+
"recall@5": 0.546079779917469,
|
1484 |
+
"recall@10": 1.0,
|
1485 |
+
"f1@1": 0.15268225584594222,
|
1486 |
+
"f1@5": 0.18202659330582308,
|
1487 |
+
"f1@10": 0.18181818181818185,
|
1488 |
+
"map@1": 0.15268225584594222,
|
1489 |
+
"map@5": 0.2876891334250344,
|
1490 |
+
"map@10": 0.3440863736599638,
|
1491 |
+
"mrr@1": 0.15268225584594222,
|
1492 |
+
"mrr@5": 0.2876891334250344,
|
1493 |
+
"mrr@10": 0.3440863736599638,
|
1494 |
+
"num_pred": 727,
|
1495 |
+
"num_data": 727
|
1496 |
+
},
|
1497 |
+
"MomentSeeker": {
|
1498 |
+
"hit@1": 0.477,
|
1499 |
+
"hit@5": 0.894,
|
1500 |
+
"hit@10": 1.0,
|
1501 |
+
"ndcg@1": 0.477,
|
1502 |
+
"ndcg@5": 0.6999431394324539,
|
1503 |
+
"ndcg@10": 0.7367693900247171,
|
1504 |
+
"precision@1": 0.477,
|
1505 |
+
"precision@5": 0.18720000000000003,
|
1506 |
+
"precision@10": 0.1063,
|
1507 |
+
"recall@1": 0.4622083333333334,
|
1508 |
+
"recall@5": 0.887375,
|
1509 |
+
"recall@10": 0.9995,
|
1510 |
+
"f1@1": 0.4665269841269841,
|
1511 |
+
"f1@5": 0.3056558996558997,
|
1512 |
+
"f1@10": 0.19023737373737373,
|
1513 |
+
"map@1": 0.477,
|
1514 |
+
"map@5": 0.6349949999999999,
|
1515 |
+
"map@10": 0.6510287698412699,
|
1516 |
+
"mrr@1": 0.477,
|
1517 |
+
"mrr@5": 0.6393833333333333,
|
1518 |
+
"mrr@10": 0.6538027777777777,
|
1519 |
+
"num_pred": 1000,
|
1520 |
+
"num_data": 1000
|
1521 |
+
},
|
1522 |
+
"K700": {
|
1523 |
+
"hit@1": 0.241,
|
1524 |
+
"hit@5": 0.525,
|
1525 |
+
"hit@10": 0.617,
|
1526 |
+
"ndcg@1": 0.241,
|
1527 |
+
"ndcg@5": 0.39440742981012006,
|
1528 |
+
"ndcg@10": 0.42430485366243514,
|
1529 |
+
"precision@1": 0.241,
|
1530 |
+
"precision@5": 0.10499999999999998,
|
1531 |
+
"precision@10": 0.0617,
|
1532 |
+
"recall@1": 0.241,
|
1533 |
+
"recall@5": 0.525,
|
1534 |
+
"recall@10": 0.617,
|
1535 |
+
"f1@1": 0.241,
|
1536 |
+
"f1@5": 0.17500000000000004,
|
1537 |
+
"f1@10": 0.1121818181818182,
|
1538 |
+
"map@1": 0.241,
|
1539 |
+
"map@5": 0.35076666666666667,
|
1540 |
+
"map@10": 0.3631940476190476,
|
1541 |
+
"mrr@1": 0.241,
|
1542 |
+
"mrr@5": 0.35076666666666667,
|
1543 |
+
"mrr@10": 0.3631940476190476,
|
1544 |
+
"num_pred": 1000,
|
1545 |
+
"num_data": 1000
|
1546 |
+
},
|
1547 |
+
"SmthSmthV2": {
|
1548 |
+
"hit@1": 0.371,
|
1549 |
+
"hit@5": 0.651,
|
1550 |
+
"hit@10": 0.753,
|
1551 |
+
"ndcg@1": 0.371,
|
1552 |
+
"ndcg@5": 0.5193728940871313,
|
1553 |
+
"ndcg@10": 0.5523936932069603,
|
1554 |
+
"precision@1": 0.371,
|
1555 |
+
"precision@5": 0.13019999999999998,
|
1556 |
+
"precision@10": 0.07529999999999999,
|
1557 |
+
"recall@1": 0.371,
|
1558 |
+
"recall@5": 0.651,
|
1559 |
+
"recall@10": 0.753,
|
1560 |
+
"f1@1": 0.371,
|
1561 |
+
"f1@5": 0.21700000000000005,
|
1562 |
+
"f1@10": 0.13690909090909092,
|
1563 |
+
"map@1": 0.371,
|
1564 |
+
"map@5": 0.4756,
|
1565 |
+
"map@10": 0.4892472222222223,
|
1566 |
+
"mrr@1": 0.371,
|
1567 |
+
"mrr@5": 0.4756,
|
1568 |
+
"mrr@10": 0.4892472222222223,
|
1569 |
+
"num_pred": 1000,
|
1570 |
+
"num_data": 1000
|
1571 |
+
},
|
1572 |
+
"HMDB51": {
|
1573 |
+
"hit@1": 0.398,
|
1574 |
+
"hit@5": 0.668,
|
1575 |
+
"hit@10": 0.75,
|
1576 |
+
"ndcg@1": 0.398,
|
1577 |
+
"ndcg@5": 0.5459200010711295,
|
1578 |
+
"ndcg@10": 0.572371610722454,
|
1579 |
+
"precision@1": 0.398,
|
1580 |
+
"precision@5": 0.1336,
|
1581 |
+
"precision@10": 0.075,
|
1582 |
+
"recall@1": 0.398,
|
1583 |
+
"recall@5": 0.668,
|
1584 |
+
"recall@10": 0.75,
|
1585 |
+
"f1@1": 0.398,
|
1586 |
+
"f1@5": 0.2226666666666667,
|
1587 |
+
"f1@10": 0.13636363636363638,
|
1588 |
+
"map@1": 0.398,
|
1589 |
+
"map@5": 0.50495,
|
1590 |
+
"map@10": 0.5158285714285714,
|
1591 |
+
"mrr@1": 0.398,
|
1592 |
+
"mrr@5": 0.50495,
|
1593 |
+
"mrr@10": 0.5158285714285714,
|
1594 |
+
"num_pred": 1000,
|
1595 |
+
"num_data": 1000
|
1596 |
+
},
|
1597 |
+
"UCF101": {
|
1598 |
+
"hit@1": 0.602,
|
1599 |
+
"hit@5": 0.847,
|
1600 |
+
"hit@10": 0.88,
|
1601 |
+
"ndcg@1": 0.602,
|
1602 |
+
"ndcg@5": 0.7371850834474476,
|
1603 |
+
"ndcg@10": 0.7477759962834329,
|
1604 |
+
"precision@1": 0.602,
|
1605 |
+
"precision@5": 0.1694,
|
1606 |
+
"precision@10": 0.088,
|
1607 |
+
"recall@1": 0.602,
|
1608 |
+
"recall@5": 0.847,
|
1609 |
+
"recall@10": 0.88,
|
1610 |
+
"f1@1": 0.602,
|
1611 |
+
"f1@5": 0.2823333333333334,
|
1612 |
+
"f1@10": 0.16,
|
1613 |
+
"map@1": 0.602,
|
1614 |
+
"map@5": 0.7001666666666667,
|
1615 |
+
"map@10": 0.7044928571428571,
|
1616 |
+
"mrr@1": 0.602,
|
1617 |
+
"mrr@5": 0.7001666666666667,
|
1618 |
+
"mrr@10": 0.7044928571428571,
|
1619 |
+
"num_pred": 1000,
|
1620 |
+
"num_data": 1000
|
1621 |
+
},
|
1622 |
+
"Breakfast": {
|
1623 |
+
"hit@1": 0.19861431870669746,
|
1624 |
+
"hit@5": 0.6882217090069284,
|
1625 |
+
"hit@10": 1.0,
|
1626 |
+
"ndcg@1": 0.19861431870669746,
|
1627 |
+
"ndcg@5": 0.43913047772660696,
|
1628 |
+
"ndcg@10": 0.5424251959619197,
|
1629 |
+
"precision@1": 0.19861431870669746,
|
1630 |
+
"precision@5": 0.13764434180138566,
|
1631 |
+
"precision@10": 0.1,
|
1632 |
+
"recall@1": 0.19861431870669746,
|
1633 |
+
"recall@5": 0.6882217090069284,
|
1634 |
+
"recall@10": 1.0,
|
1635 |
+
"f1@1": 0.19861431870669746,
|
1636 |
+
"f1@5": 0.2294072363356429,
|
1637 |
+
"f1@10": 0.1818181818181818,
|
1638 |
+
"map@1": 0.19861431870669746,
|
1639 |
+
"map@5": 0.35816012317167056,
|
1640 |
+
"map@10": 0.4022590637486711,
|
1641 |
+
"mrr@1": 0.19861431870669746,
|
1642 |
+
"mrr@5": 0.35816012317167056,
|
1643 |
+
"mrr@10": 0.4022590637486711,
|
1644 |
+
"num_pred": 433,
|
1645 |
+
"num_data": 433
|
1646 |
+
},
|
1647 |
+
"MVBench": "FILE_N/A",
|
1648 |
+
"Video-MME": {
|
1649 |
+
"hit@1": 0.2837037037037037,
|
1650 |
+
"hit@5": 1.0,
|
1651 |
+
"hit@10": 1.0,
|
1652 |
+
"ndcg@1": 0.2837037037037037,
|
1653 |
+
"ndcg@5": 0.6635977117783535,
|
1654 |
+
"ndcg@10": 0.6635977117783535,
|
1655 |
+
"precision@1": 0.2837037037037037,
|
1656 |
+
"precision@5": 0.19999999999999996,
|
1657 |
+
"precision@10": 0.09999999999999998,
|
1658 |
+
"recall@1": 0.2837037037037037,
|
1659 |
+
"recall@5": 1.0,
|
1660 |
+
"recall@10": 1.0,
|
1661 |
+
"f1@1": 0.2837037037037037,
|
1662 |
+
"f1@5": 0.3333333333333334,
|
1663 |
+
"f1@10": 0.18181818181818188,
|
1664 |
+
"map@1": 0.2837037037037037,
|
1665 |
+
"map@5": 0.5511728395061728,
|
1666 |
+
"map@10": 0.5511728395061728,
|
1667 |
+
"mrr@1": 0.2837037037037037,
|
1668 |
+
"mrr@5": 0.5511728395061728,
|
1669 |
+
"mrr@10": 0.5511728395061728,
|
1670 |
+
"num_pred": 2700,
|
1671 |
+
"num_data": 2700
|
1672 |
+
},
|
1673 |
+
"NExTQA": {
|
1674 |
+
"hit@1": 0.23505371321812238,
|
1675 |
+
"hit@5": 1.0,
|
1676 |
+
"hit@10": 1.0,
|
1677 |
+
"ndcg@1": 0.23505371321812238,
|
1678 |
+
"ndcg@5": 0.6269658989529459,
|
1679 |
+
"ndcg@10": 0.6269658989529459,
|
1680 |
+
"precision@1": 0.23505371321812238,
|
1681 |
+
"precision@5": 0.20000000000000004,
|
1682 |
+
"precision@10": 0.10000000000000002,
|
1683 |
+
"recall@1": 0.23505371321812238,
|
1684 |
+
"recall@5": 1.0,
|
1685 |
+
"recall@10": 1.0,
|
1686 |
+
"f1@1": 0.23505371321812238,
|
1687 |
+
"f1@5": 0.3333333333333333,
|
1688 |
+
"f1@10": 0.1818181818181818,
|
1689 |
+
"map@1": 0.23505371321812238,
|
1690 |
+
"map@5": 0.5042367273859567,
|
1691 |
+
"map@10": 0.5042367273859567,
|
1692 |
+
"mrr@1": 0.23505371321812238,
|
1693 |
+
"mrr@5": 0.5040732523742799,
|
1694 |
+
"mrr@10": 0.5040732523742799,
|
1695 |
+
"num_pred": 8564,
|
1696 |
+
"num_data": 8564
|
1697 |
+
},
|
1698 |
+
"EgoSchema": "FILE_N/A",
|
1699 |
+
"ActivityNetQA": "FILE_N/A"
|
1700 |
+
}
|
1701 |
+
}
|
1702 |
+
}
|
app.py
CHANGED
@@ -1,4 +1,5 @@
|
|
1 |
from utils import *
|
|
|
2 |
|
3 |
global data_component
|
4 |
|
@@ -10,12 +11,20 @@ def update_table(query, min_size, max_size, selected_tasks=None):
|
|
10 |
filtered_df = filtered_df[selected_columns]
|
11 |
return filtered_df
|
12 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
with gr.Blocks() as block:
|
14 |
gr.Markdown(LEADERBOARD_INTRODUCTION)
|
15 |
|
16 |
with gr.Tabs(elem_classes="tab-buttons") as tabs:
|
17 |
# Table 1
|
18 |
-
with gr.TabItem("📊 MMEB", elem_id="qa-tab-table1", id=1):
|
19 |
with gr.Row():
|
20 |
with gr.Accordion("Citation", open=False):
|
21 |
citation_button = gr.Textbox(
|
@@ -54,7 +63,7 @@ with gr.Blocks() as block:
|
|
54 |
|
55 |
with gr.Row():
|
56 |
tasks_select = gr.CheckboxGroup(
|
57 |
-
choices=TASKS_V1
|
58 |
value=TASKS_V1,
|
59 |
label="Select tasks to Display",
|
60 |
elem_id="tasks-select"
|
@@ -97,13 +106,95 @@ with gr.Blocks() as block:
|
|
97 |
)
|
98 |
refresh_button.click(fn=refresh_data, outputs=data_component)
|
99 |
|
100 |
-
|
101 |
-
|
102 |
-
|
103 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
104 |
|
105 |
# table 3
|
106 |
-
with gr.TabItem("
|
|
|
|
|
|
|
|
|
|
|
107 |
with gr.Row():
|
108 |
gr.Markdown(SUBMIT_INTRODUCTION, elem_classes="markdown-text")
|
109 |
|
|
|
1 |
from utils import *
|
2 |
+
import utils_v2 as v2
|
3 |
|
4 |
global data_component
|
5 |
|
|
|
11 |
filtered_df = filtered_df[selected_columns]
|
12 |
return filtered_df
|
13 |
|
14 |
+
def update_table_v2(query, min_size, max_size, selected_tasks=None):
|
15 |
+
df = v2.get_df()
|
16 |
+
filtered_df = v2.search_and_filter_models(df, query, min_size, max_size)
|
17 |
+
if selected_tasks and len(selected_tasks) > 0:
|
18 |
+
selected_columns = v2.BASE_COLS + selected_tasks
|
19 |
+
filtered_df = filtered_df[selected_columns]
|
20 |
+
return filtered_df
|
21 |
+
|
22 |
with gr.Blocks() as block:
|
23 |
gr.Markdown(LEADERBOARD_INTRODUCTION)
|
24 |
|
25 |
with gr.Tabs(elem_classes="tab-buttons") as tabs:
|
26 |
# Table 1
|
27 |
+
with gr.TabItem("📊 MMEB-V1", elem_id="qa-tab-table1", id=1):
|
28 |
with gr.Row():
|
29 |
with gr.Accordion("Citation", open=False):
|
30 |
citation_button = gr.Textbox(
|
|
|
63 |
|
64 |
with gr.Row():
|
65 |
tasks_select = gr.CheckboxGroup(
|
66 |
+
choices=TASKS_V1,
|
67 |
value=TASKS_V1,
|
68 |
label="Select tasks to Display",
|
69 |
elem_id="tasks-select"
|
|
|
106 |
)
|
107 |
refresh_button.click(fn=refresh_data, outputs=data_component)
|
108 |
|
109 |
+
with gr.TabItem("📊 MMEB-V2", elem_id="qa-tab-table1", id=2):
|
110 |
+
with gr.Row():
|
111 |
+
with gr.Accordion("Citation", open=False):
|
112 |
+
citation_button2 = gr.Textbox(
|
113 |
+
value=v2.CITATION_BUTTON_TEXT,
|
114 |
+
label=CITATION_BUTTON_LABEL,
|
115 |
+
elem_id="citation-button",
|
116 |
+
lines=10,
|
117 |
+
)
|
118 |
+
gr.Markdown(v2.TABLE_INTRODUCTION)
|
119 |
+
|
120 |
+
with gr.Row():
|
121 |
+
search_bar2 = gr.Textbox(
|
122 |
+
placeholder="Search models...",
|
123 |
+
show_label=False,
|
124 |
+
elem_id="search-bar"
|
125 |
+
)
|
126 |
+
|
127 |
+
df2 = v2.get_df()
|
128 |
+
min_size2, max_size2 = get_size_range(df2)
|
129 |
+
|
130 |
+
with gr.Row():
|
131 |
+
min_size_slider2 = gr.Slider(
|
132 |
+
minimum=min_size2,
|
133 |
+
maximum=max_size2,
|
134 |
+
value=min_size2,
|
135 |
+
step=0.1,
|
136 |
+
label="Minimum number of parameters (B)",
|
137 |
+
)
|
138 |
+
max_size_slider2 = gr.Slider(
|
139 |
+
minimum=min_size2,
|
140 |
+
maximum=max_size2,
|
141 |
+
value=max_size2,
|
142 |
+
step=0.1,
|
143 |
+
label="Maximum number of parameters (B)",
|
144 |
+
)
|
145 |
+
|
146 |
+
with gr.Row():
|
147 |
+
tasks_select2 = gr.CheckboxGroup(
|
148 |
+
choices=v2.TASKS,
|
149 |
+
value=v2.TASKS,
|
150 |
+
label="Select tasks to Display",
|
151 |
+
elem_id="tasks-select"
|
152 |
+
)
|
153 |
+
|
154 |
+
data_component2 = gr.components.Dataframe(
|
155 |
+
value=df2[v2.COLUMN_NAMES],
|
156 |
+
headers=v2.COLUMN_NAMES,
|
157 |
+
type="pandas",
|
158 |
+
datatype=v2.DATA_TITLE_TYPE,
|
159 |
+
interactive=False,
|
160 |
+
visible=True,
|
161 |
+
max_height=2400,
|
162 |
+
)
|
163 |
+
|
164 |
+
refresh_button2 = gr.Button("Refresh")
|
165 |
+
|
166 |
+
def update_with_tasks_v2(*args):
|
167 |
+
return update_table_v2(*args)
|
168 |
+
|
169 |
+
search_bar2.change(
|
170 |
+
fn=update_with_tasks_v2,
|
171 |
+
inputs=[search_bar2, min_size_slider2, max_size_slider2, tasks_select2],
|
172 |
+
outputs=data_component2
|
173 |
+
)
|
174 |
+
min_size_slider2.change(
|
175 |
+
fn=update_with_tasks_v2,
|
176 |
+
inputs=[search_bar2, min_size_slider2, max_size_slider2, tasks_select2],
|
177 |
+
outputs=data_component2
|
178 |
+
)
|
179 |
+
max_size_slider2.change(
|
180 |
+
fn=update_with_tasks_v2,
|
181 |
+
inputs=[search_bar2, min_size_slider2, max_size_slider2, tasks_select2],
|
182 |
+
outputs=data_component2
|
183 |
+
)
|
184 |
+
tasks_select2.change(
|
185 |
+
fn=update_with_tasks_v2,
|
186 |
+
inputs=[search_bar2, min_size_slider2, max_size_slider2, tasks_select2],
|
187 |
+
outputs=data_component2
|
188 |
+
)
|
189 |
+
refresh_button.click(fn=v2.refresh_data, outputs=data_component)
|
190 |
|
191 |
# table 3
|
192 |
+
with gr.TabItem("📝 About", elem_id="qa-tab-table2", id=3):
|
193 |
+
gr.Markdown(LEADERBOARD_INFO, elem_classes="markdown-text")
|
194 |
+
# gr.Image("overview.png", width=900, label="Dataset Overview")
|
195 |
+
|
196 |
+
# table 4
|
197 |
+
with gr.TabItem("🚀 Submit here! ", elem_id="submit-tab", id=4):
|
198 |
with gr.Row():
|
199 |
gr.Markdown(SUBMIT_INTRODUCTION, elem_classes="markdown-text")
|
200 |
|
utils.py
CHANGED
@@ -12,14 +12,13 @@ HF_TOKEN = os.environ.get("HF_TOKEN")
|
|
12 |
|
13 |
BASE_COLS = ["Rank", "Models", "Model Size(B)", "Data Source"]
|
14 |
TASKS_V1 = ["V1-Overall", "I-CLS", "I-QA", "I-RET", "I-VG"]
|
15 |
-
|
16 |
-
COLUMN_NAMES = BASE_COLS + TASKS_V1 + TASKS_V2
|
17 |
|
18 |
DATA_TITLE_TYPE = ['number', 'markdown', 'str', 'markdown'] + \
|
19 |
-
['number'] * len(TASKS_V1
|
20 |
|
21 |
LEADERBOARD_INTRODUCTION = """
|
22 |
-
# 📊 **MMEB LEADERBOARD
|
23 |
|
24 |
## Introduction
|
25 |
We introduce a novel benchmark, **MMEB-V1 (Massive Multimodal Embedding Benchmark)**,
|
@@ -59,7 +58,7 @@ SUBMIT_INTRODUCTION = """# Submit on MMEB Leaderboard Introduction
|
|
59 |
|
60 |
## ⚠ Please note that you need to submit the JSON file with the following format:
|
61 |
|
62 |
-
### **TO SUBMIT V1 ONLY**
|
63 |
```json
|
64 |
[
|
65 |
{
|
@@ -76,26 +75,56 @@ SUBMIT_INTRODUCTION = """# Submit on MMEB Leaderboard Introduction
|
|
76 |
]
|
77 |
```
|
78 |
|
79 |
-
###
|
|
|
|
|
80 |
```json
|
81 |
-
|
82 |
-
{
|
83 |
"Model": "<Model Name>",
|
84 |
"URL": "<Model URL>" or null,
|
85 |
"Model Size(B)": 1000 or null,
|
86 |
"Data Source": "Self-Reported",
|
87 |
-
"
|
88 |
-
"
|
89 |
-
|
90 |
-
|
91 |
-
"
|
92 |
-
|
93 |
-
|
94 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
95 |
```
|
96 |
-
You are also welcome to submit both versions by including all the fields above! :) \n
|
97 |
You may refer to the [**GitHub page**](https://github.com/TIGER-AI-Lab/VLM2Vec) for instructions about evaluating your model. \n
|
98 |
-
Please send us an email at [email protected], attaching the JSON file. We will review your submission and update the leaderboard accordingly.
|
|
|
99 |
"""
|
100 |
|
101 |
def create_hyperlinked_names(df):
|
@@ -124,7 +153,7 @@ def create_hyperlinked_names(df):
|
|
124 |
def get_df(file="results.jsonl"):
|
125 |
df = pd.read_json(file, orient='records', lines=True)
|
126 |
df['Model Size(B)'] = df['Model Size(B)'].apply(process_model_size)
|
127 |
-
for task in TASKS_V1
|
128 |
if df[task].isnull().any():
|
129 |
df[task] = df[task].apply(lambda score: '-' if pd.isna(score) else score)
|
130 |
df = df.sort_values(by=['V1-Overall'], ascending=False)
|
|
|
12 |
|
13 |
BASE_COLS = ["Rank", "Models", "Model Size(B)", "Data Source"]
|
14 |
TASKS_V1 = ["V1-Overall", "I-CLS", "I-QA", "I-RET", "I-VG"]
|
15 |
+
COLUMN_NAMES = BASE_COLS + TASKS_V1
|
|
|
16 |
|
17 |
DATA_TITLE_TYPE = ['number', 'markdown', 'str', 'markdown'] + \
|
18 |
+
['number'] * len(TASKS_V1)
|
19 |
|
20 |
LEADERBOARD_INTRODUCTION = """
|
21 |
+
# 📊 **MMEB LEADERBOARD**
|
22 |
|
23 |
## Introduction
|
24 |
We introduce a novel benchmark, **MMEB-V1 (Massive Multimodal Embedding Benchmark)**,
|
|
|
58 |
|
59 |
## ⚠ Please note that you need to submit the JSON file with the following format:
|
60 |
|
61 |
+
### **TO SUBMIT V1 ONLY (Depreciating)**
|
62 |
```json
|
63 |
[
|
64 |
{
|
|
|
75 |
]
|
76 |
```
|
77 |
|
78 |
+
### ***Important Notes: We will be releasing MMEB-V2 soon!***
|
79 |
+
### ***In V2, the detailed scores of each dataset will be included, and our code will automatically generate the results and calculate the overall scores.***
|
80 |
+
### **A V2 Submission would look like this: (TO BE RELEASED SOON)**
|
81 |
```json
|
82 |
+
{
|
83 |
+
"metadata": {
|
84 |
"Model": "<Model Name>",
|
85 |
"URL": "<Model URL>" or null,
|
86 |
"Model Size(B)": 1000 or null,
|
87 |
"Data Source": "Self-Reported",
|
88 |
+
"V1-Overall": 50.0,
|
89 |
+
"V2-Overall": 50.0
|
90 |
+
},
|
91 |
+
"metrics": {
|
92 |
+
"image": {
|
93 |
+
"ImageNet-1K": {
|
94 |
+
"hit@1": 0.5,
|
95 |
+
"ndcg@1": 0.5,
|
96 |
+
... ...
|
97 |
+
},
|
98 |
+
"N24News": {
|
99 |
+
... ...
|
100 |
+
},
|
101 |
+
... ...
|
102 |
+
},
|
103 |
+
"visdoc": {
|
104 |
+
"ViDoRe": {
|
105 |
+
"hit@1": 0.5,
|
106 |
+
"ndcg@1": 0.5,
|
107 |
+
... ...
|
108 |
+
},
|
109 |
+
... ...
|
110 |
+
},
|
111 |
+
"video": {
|
112 |
+
"DiDeMo": {
|
113 |
+
"hit@1": 0.5,
|
114 |
+
"ndcg@1": 0.5,
|
115 |
+
... ...
|
116 |
+
},
|
117 |
+
"MSR-VTT": {
|
118 |
+
... ...
|
119 |
+
},
|
120 |
+
... ...
|
121 |
+
}
|
122 |
+
}
|
123 |
+
}
|
124 |
```
|
|
|
125 |
You may refer to the [**GitHub page**](https://github.com/TIGER-AI-Lab/VLM2Vec) for instructions about evaluating your model. \n
|
126 |
+
Please send us an email at [email protected], attaching the JSON file. We will review your submission and update the leaderboard accordingly. \n
|
127 |
+
Please also share any feedback or suggestions you have for improving the leaderboard experience. We appreciate your contributions to the MMEB community!
|
128 |
"""
|
129 |
|
130 |
def create_hyperlinked_names(df):
|
|
|
153 |
def get_df(file="results.jsonl"):
|
154 |
df = pd.read_json(file, orient='records', lines=True)
|
155 |
df['Model Size(B)'] = df['Model Size(B)'].apply(process_model_size)
|
156 |
+
for task in TASKS_V1:
|
157 |
if df[task].isnull().any():
|
158 |
df[task] = df[task].apply(lambda score: '-' if pd.isna(score) else score)
|
159 |
df = df.sort_values(by=['V1-Overall'], ascending=False)
|
utils_v2.py
ADDED
@@ -0,0 +1,149 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import json
|
2 |
+
import os
|
3 |
+
import pandas as pd
|
4 |
+
from utils import create_hyperlinked_names
|
5 |
+
|
6 |
+
def sum_lst(lst):
|
7 |
+
assert isinstance(lst, list) and lst, f"Input should be a non-empty list, got {type(lst)}, size {len(lst)}"
|
8 |
+
total = lst[0]
|
9 |
+
for item in lst[1:]:
|
10 |
+
assert isinstance(item, (list, int, float)), f"Expected types are list and numbers, got {type(item)}"
|
11 |
+
total += item
|
12 |
+
return total
|
13 |
+
|
14 |
+
SCORE_BASE_DIR = "Scores"
|
15 |
+
META_DATA = ["model_name", "model_size", "url"]
|
16 |
+
DATASETS = {
|
17 |
+
"image": {
|
18 |
+
"I-CLS": ['VOC2007', 'N24News', 'SUN397', 'ObjectNet', 'Country211', 'Place365', 'ImageNet-1K', 'HatefulMemes', 'ImageNet-A', 'ImageNet-R'],
|
19 |
+
"I-QA": ['OK-VQA', 'A-OKVQA', 'DocVQA', 'InfographicsVQA', 'ChartQA', 'Visual7W-Pointing', 'ScienceQA', 'GQA', 'TextVQA', 'VizWiz'],
|
20 |
+
"I-RET": ['VisDial', 'CIRR', 'VisualNews_t2i', 'VisualNews_i2t', 'MSCOCO_t2i', 'MSCOCO_i2t', 'NIGHTS', 'WebQA', 'FashionIQ', 'Wiki-SS-NQ', 'OVEN', 'EDIS'],
|
21 |
+
"I-VG": ['MSCOCO', 'RefCOCO', 'RefCOCO-Matching', 'Visual7W']
|
22 |
+
},
|
23 |
+
"visdoc": {
|
24 |
+
"VisDoc": ['ViDoRe_arxivqa', 'ViDoRe_docvqa', 'ViDoRe_infovqa', 'ViDoRe_tabfquad', 'ViDoRe_tatdqa', 'ViDoRe_shiftproject', 'ViDoRe_syntheticDocQA_artificial_intelligence', 'ViDoRe_syntheticDocQA_energy', 'ViDoRe_syntheticDocQA_government_reports', 'ViDoRe_syntheticDocQA_healthcare_industry', 'VisRAG_ArxivQA', 'VisRAG_ChartQA', 'VisRAG_MP-DocVQA', 'VisRAG_SlideVQA', 'VisRAG_InfoVQA', 'VisRAG_PlotQA', 'ViDoSeek-page', 'ViDoSeek-doc', 'MMLongBench-page', 'MMLongBench-doc']
|
25 |
+
},
|
26 |
+
"video": {
|
27 |
+
"V-CLS": ['K700', 'UCF101', 'HMDB51', 'SmthSmthV2', 'Breakfast'],
|
28 |
+
"V-QA": ['Video-MME', 'MVBench', 'NExTQA', 'EgoSchema'],
|
29 |
+
"V-RET": ['MSR-VTT', 'MSVD', 'DiDeMo', 'VATEX', 'YouCook2'],
|
30 |
+
"V-MRET": ['QVHighlight', 'Charades-STA', 'MomentSeeker', 'ActivityNetQA']
|
31 |
+
}
|
32 |
+
}
|
33 |
+
ALL_DATASETS_SPLITS = {k: sum_lst(list(v.values())) for k, v in DATASETS.items()}
|
34 |
+
ALL_DATASETS = sum_lst(list(ALL_DATASETS_SPLITS.values()))
|
35 |
+
MODALITIES = list(DATASETS.keys())
|
36 |
+
SPECIAL_METRICS = {
|
37 |
+
'__default__': 'hit@1',
|
38 |
+
}
|
39 |
+
|
40 |
+
BASE_COLS = ['Rank', 'Models', 'Model Size(B)']
|
41 |
+
TASKS = ["Overall", "Image-Overall", "I-CLS", "I-QA", "I-RET", "I-VG", "VisDoc", "Video-Overall", "V-CLS", "V-QA", "V-RET", "V-MRET"]
|
42 |
+
COLUMN_NAMES = BASE_COLS + TASKS
|
43 |
+
|
44 |
+
DATA_TITLE_TYPE = ['number', 'markdown', 'str', 'markdown'] + \
|
45 |
+
['number'] * len(TASKS)
|
46 |
+
|
47 |
+
TABLE_INTRODUCTION = """"""
|
48 |
+
|
49 |
+
LEADERBOARD_INFO = """
|
50 |
+
## Dataset Summary
|
51 |
+
"""
|
52 |
+
|
53 |
+
CITATION_BUTTON_TEXT = r""""""
|
54 |
+
|
55 |
+
def load_single_json(file_path):
|
56 |
+
with open(file_path, 'r') as file:
|
57 |
+
data = json.load(file)
|
58 |
+
return data
|
59 |
+
|
60 |
+
def load_data(base_dir=SCORE_BASE_DIR):
|
61 |
+
all_data = []
|
62 |
+
for file_name in os.listdir(base_dir):
|
63 |
+
if file_name.endswith('-scores_report.json'):
|
64 |
+
file_path = os.path.join(base_dir, file_name)
|
65 |
+
data = load_single_json(file_path)
|
66 |
+
all_data.append(data)
|
67 |
+
return all_data
|
68 |
+
|
69 |
+
def calculate_score(raw_scores=None):
|
70 |
+
"""This function calculates the overall average scores for all datasets as well as avg scores for each modality and sub-task based on the raw scores.
|
71 |
+
Algorithm summary:
|
72 |
+
"""
|
73 |
+
def get_avg(sum_score, leng):
|
74 |
+
avg = sum_score / leng if leng > 0 else 0.0
|
75 |
+
avg = round(avg, 2) # Round to 2 decimal places
|
76 |
+
return avg
|
77 |
+
|
78 |
+
avg_scores = {}
|
79 |
+
overall_scores_summary = {} # Stores the scores sum and length for each modality and all datasets
|
80 |
+
for modality, datasets_list in DATASETS.items(): # Ex.: ('image', {'I-CLS': [...], 'I-QA': [...]})
|
81 |
+
overall_scores_summary[modality] = (0.0, 0) # Initialize the sum and count for each modality
|
82 |
+
for sub_task, datasets in datasets_list.items(): # Ex.: ('I-CLS', ['VOC2007', 'N24News', ...])
|
83 |
+
sub_task_sum_score, sub_task_datasets_len = 0.0, len(datasets)
|
84 |
+
for dataset in datasets: # Ex.: 'VOC2007'
|
85 |
+
score = raw_scores.get(modality, {}).get(dataset, 0.0)
|
86 |
+
score = 0.0 if score == "FILE_N/A" else score
|
87 |
+
metric = SPECIAL_METRICS.get(dataset, 'hit@1')
|
88 |
+
if isinstance(score, dict):
|
89 |
+
score = score.get(metric, 0.0)
|
90 |
+
sub_task_sum_score += score
|
91 |
+
|
92 |
+
sub_task_overall = get_avg(sub_task_sum_score, sub_task_datasets_len)
|
93 |
+
avg_scores[sub_task] = sub_task_overall
|
94 |
+
|
95 |
+
# Accumulate the scores sum and length for the each modality
|
96 |
+
modality_sum_score, modality_datasets_len = overall_scores_summary[modality]
|
97 |
+
modality_sum_score += sub_task_sum_score
|
98 |
+
modality_datasets_len += sub_task_datasets_len
|
99 |
+
overall_scores_summary[modality] = (modality_sum_score, modality_datasets_len)
|
100 |
+
|
101 |
+
all_datasets_sum_score, all_datasets_len = 0.0, 0
|
102 |
+
for modality, (modality_sum_score, modality_datasets_len) in overall_scores_summary.items():
|
103 |
+
name = f"{modality.capitalize()}-Overall"
|
104 |
+
avg_scores[name] = get_avg(modality_sum_score, modality_datasets_len)
|
105 |
+
# Accumulate the scores sum and length for all datasets
|
106 |
+
all_datasets_sum_score += modality_sum_score
|
107 |
+
all_datasets_len += modality_datasets_len
|
108 |
+
avg_scores['Overall'] = get_avg(all_datasets_sum_score, all_datasets_len)
|
109 |
+
return avg_scores
|
110 |
+
|
111 |
+
def generate_model_row(data):
|
112 |
+
metadata = data['metadata']
|
113 |
+
row = {
|
114 |
+
'Models': metadata.get('model_name', None),
|
115 |
+
'Model Size(B)': metadata.get('model_size', None),
|
116 |
+
'URL': metadata.get('url', None)
|
117 |
+
}
|
118 |
+
scores = calculate_score(data['metrics'])
|
119 |
+
row.update(scores)
|
120 |
+
return row
|
121 |
+
|
122 |
+
def get_df():
|
123 |
+
"""Generates a DataFrame from the loaded data."""
|
124 |
+
all_data = load_data()
|
125 |
+
rows = [generate_model_row(data) for data in all_data]
|
126 |
+
df = pd.DataFrame(rows)
|
127 |
+
df = df.sort_values(by='Overall', ascending=False).reset_index(drop=True)
|
128 |
+
df['Rank'] = range(1, len(df) + 1)
|
129 |
+
df = create_hyperlinked_names(df)
|
130 |
+
|
131 |
+
return df
|
132 |
+
|
133 |
+
def refresh_data():
|
134 |
+
df = get_df()
|
135 |
+
return df[COLUMN_NAMES]
|
136 |
+
|
137 |
+
def search_and_filter_models(df, query, min_size, max_size):
|
138 |
+
filtered_df = df.copy()
|
139 |
+
|
140 |
+
if query:
|
141 |
+
filtered_df = filtered_df[filtered_df['Models'].str.contains(query, case=False, na=False)]
|
142 |
+
|
143 |
+
size_mask = filtered_df['Model Size(B)'].apply(lambda x:
|
144 |
+
(min_size <= 1000.0 <= max_size) if x == 'unknown'
|
145 |
+
else (min_size <= x <= max_size))
|
146 |
+
|
147 |
+
filtered_df = filtered_df[size_mask]
|
148 |
+
|
149 |
+
return filtered_df[COLUMN_NAMES]
|